Sentaurus™ Mesh User

Guide
Version O-2018.06, June 2018
Copyright and Proprietary Information Notice
© 2018 Synopsys, Inc. This Synopsys software and all associated documentation are proprietary to Synopsys, Inc. and may only be
used pursuant to the terms and conditions of a written license agreement with Synopsys, Inc. All other use, reproduction,
modification, or distribution of the Synopsys software or the associated documentation is strictly prohibited.
Destination Control Statement
All technical data contained in this publication is subject to the export control laws of the United States of America.
Disclosure to nationals of other countries contrary to United States law is prohibited. It is the reader’s responsibility to
determine the applicable regulations and to comply with them.
Disclaimer
SYNOPSYS, INC., AND ITS LICENSORS MAKE NO WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, WITH
REGARD TO THIS MATERIAL, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE.
Trademarks
Synopsys and certain Synopsys product names are trademarks of Synopsys, as set forth at
https://www.synopsys.com/company/legal/trademarks-brands.html.
All other product or company names may be trademarks of their respective owners.
Third-Party Links
Any links to third-party websites included in this document are for your convenience only. Synopsys does not endorse
and is not responsible for such websites and their practices, including privacy practices, availability, and content.

Synopsys, Inc.
690 E. Middlefield Road
Mountain View, CA 94043
www.synopsys.com

ii Sentaurus™ Mesh User Guide

O-2018.06
 Contents
About This Guide vii
Related Publications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii
Conventions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii
Customer Support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii
Accessing SolvNet. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . viii
Contacting Synopsys Support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . viii
Contacting Your Local TCAD Support Team Directly. . . . . . . . . . . . . . . . . . . . . . . viii

Chapter 1 Introduction to Sentaurus Mesh 1

Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Applications of Different Mesh Generators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
Starting Sentaurus Mesh . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Command-Line Options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

Chapter 2 Command File 5

Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
IOControls Section . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
Definitions Section . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Defining Refinement Regions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Defining Multibox Regions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Defining Constant Profiles. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
Defining Analytic Profiles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
Specifying a Gaussian Function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
Specifying an Error Function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
Specifying a 1D External Profile . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
Using the General Function Evaluator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
Defining Submeshes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
Defining Particle Profiles. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
Placements Section . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
Geometric Elements. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
Placing Refinement Regions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
Placing Multibox Regions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
Placing Constant Profiles. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
Placing Analytic Profiles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
Placing Submeshes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

Sentaurus™ Mesh User Guide iii

O-2018.06
Contents

Placing Particle Profiles. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28

Interpolate Section . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
AxisAligned Section . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
Offsetting Section . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
Delaunizer Section . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
Delaunay Tolerance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
Tensor Section. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
Mesh Subsection for Controlling Mesh Generation . . . . . . . . . . . . . . . . . . . . . . . . . . 45
EMW Subsection for Computing Cell Size Automatically . . . . . . . . . . . . . . . . . . . . . 49
Box Subsection for Plotting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
Tools Section. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
Appending the Input Structure. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
Creating Profiles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
Setting a Transformation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
Removing Short Features. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
Rediscretizing the Boundary Using the DelPSC Algorithm . . . . . . . . . . . . . . . . . . . . 57
Rediscretizing the Boundary Using the Dual-Contouring Algorithm . . . . . . . . . . . . . 57
Interpolating a Source Mesh to a Destination Mesh . . . . . . . . . . . . . . . . . . . . . . . . . . 58
Performing a 2D Slice of 3D Mesh or Boundary . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
Cutting a Mesh With a Plane . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
Reflecting a Mesh . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
Extruding a Mesh. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
Stretching a Mesh . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
Placing Individual Dopant of Species . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
Extracting a Boundary From a Mesh . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
Converting a Tetrahedral Mesh to a Hybrid Mesh . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
Specifying Algorithm for Smoothing Noise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
Creating Structures With Randomized Doping Profiles . . . . . . . . . . . . . . . . . . . . . . . 64
Adding or Removing Interfaces From a Mesh . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
QualityReport Section. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71

Chapter 3 Doping and Refinement Examples 73

Command File for a Simple Diode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
Refinement and Evaluation Windows. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
Using Refinement Polygons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
Using Composite Elements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
Regionwise and Materialwise Refinement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
Using Analytic Functions for Doping Specification. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
Creating 3D Profiles From 2D Cross Sections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
Using Particle Profiles to Specify Doping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81

iv Sentaurus™ Mesh User Guide

O-2018.06
 Contents

Generating 2D Mesh With Continuous Doping Obtained From 3D KMC File

Containing Particle Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
Performing Interface Refinement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
Ignoring Interfaces Between Regions of the Same Material . . . . . . . . . . . . . . . . . . . . 86
Offsetting Mesh Generation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
Simple Example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
Layering From All Boundaries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
Localizing the Refinement Using Cuts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
Using Analytic Functions for Refinement I . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
Using Analytic Functions for Refinement II. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95

Chapter 4 Tensor-Product Examples 97

Simple Cube . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
Using Boundary and Command Files to Generate Doping and Refinement . . . . . . . . . . 99
Thin Regions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
Computing Cell Size Automatically (EMW Applications) . . . . . . . . . . . . . . . . . . . . . . 101

Chapter 5 Tools Section 103

Activating the Tools Section. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
Reflecting and Sweeping Mesh. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
Slicing a 3D Mesh Using a Plane and Its Location. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
Cutting a 3D Mesh . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106
Converting a Tetrahedral Mesh to a Hybrid Mesh . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
Generating Randomized Doping From Continuous Doping. . . . . . . . . . . . . . . . . . . . . . 108
Slicing a 3D Mesh Using a Segment and a Direction. . . . . . . . . . . . . . . . . . . . . . . . . . . 110
Creating Profiles in an Existing Mesh . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111
Stretching a Mesh . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112

Chapter 6 Delaunization Algorithm 113

Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113
Generating Ridges and Corners . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114
Protecting Ridges and Corners . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114
Conforming Delaunay Triangulation Algorithm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114
Optimizing Elements. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115
Eliminating Slivers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115

Sentaurus™ Mesh User Guide v

O-2018.06
Contents

Appendix A Formulas for Analytic Profiles 117

General Concepts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
Local Coordinate Systems, Valid Domains, and Reference Regions . . . . . . . . . . . . 117
One-Dimensional Profiles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
Two-Dimensional Profiles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
Three-Dimensional Profiles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
General Implantation Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
Gaussian Function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120
Error Function. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120
Other Relevant Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
Dose . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
Values at the Junction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122
Length. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122
Available Models Along the Primary Direction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122
Gaussian Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
Error Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
Constant Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125
1D External Profiles. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125
Lateral or Decay Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125
Lateral Gaussian Function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
Lateral Error Function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
No Lateral Function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128

Appendix B Doping Function for Discrete Dopants 129

Doping Function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
Cut-off Parameter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131

vi Sentaurus™ Mesh User Guide

O-2018.06
 About This Guide
The Synopsys Sentaurus™ Mesh tool is a mesh generator that incorporates different mesh
generation engines: an axis-aligned mesh generator, an offsetting mesh generator, and a tensor-
product mesh generator that produces rectangular or hexahedral elements. Sentaurus Mesh is
designed to be used in a wide range of simulators, including the Synopsys TCAD products
Sentaurus Device, Sentaurus Process, Sentaurus Device Electromagnetic Wave Solver, and
Sentaurus Interconnect. Local mesh refinement is performed using the doping and refinement
information in the mesh command file.

Related Publications
For additional information, see:
■ The TCAD Sentaurus release notes, available on the Synopsys SolvNet® support site (see
Accessing SolvNet on page viii).
■ Documentation available on SolvNet at https://solvnet.synopsys.com/DocsOnWeb.

Conventions
The following conventions are used in Synopsys documentation.

Convention Description

Blue text Identifies a cross-reference (only on the screen).

Bold text Identifies a selectable icon, button, menu, or tab. It also indicates the name of a field or an
option.

Courier font Identifies text that is displayed on the screen or that the user must type. It identifies the
names of files, directories, paths, parameters, keywords, and variables.

Italicized text Used for emphasis, the titles of books and journals, and non-English words. It also
identifies components of an equation or a formula, a placeholder, or an identifier.

Customer Support
Customer support is available through the Synopsys SolvNet customer support website and by
contacting the Synopsys support center.

Sentaurus™ Mesh User Guide vii

O-2018.06
About This Guide
Customer Support

Accessing SolvNet
The SolvNet support site includes an electronic knowledge base of technical articles and
answers to frequently asked questions about Synopsys tools. The site also gives you access to
a wide range of Synopsys online services, which include downloading software, viewing
documentation, and entering a call to the Support Center.

To access the SolvNet site:

1. Go to the web page at https://solvnet.synopsys.com.
2. If prompted, enter your user name and password. (If you do not have a Synopsys user name
and password, follow the instructions to register.)

If you need help using the site, click Help on the menu bar.

Contacting Synopsys Support

If you have problems, questions, or suggestions, you can contact Synopsys support in the
following ways:
■ Go to the Synopsys Global Support Centers site on synopsys.com. There you can find e-
mail addresses and telephone numbers for Synopsys support centers throughout the world.
■ Go to either the Synopsys SolvNet site or the Synopsys Global Support Centers site and
open a case online (Synopsys user name and password required).

Contacting Your Local TCAD Support Team Directly

Send an e-mail message to:
■ [email protected] from within North America and South America
■ [email protected] from within Europe
■ [email protected] from within Asia Pacific (China, Taiwan, Singapore,
Malaysia, India, Australia)
■ [email protected] from Korea
■ [email protected] from Japan

viii Sentaurus™ Mesh User Guide

O-2018.06
CHAPTER 1 Introduction to Sentaurus Mesh

This chapter describes how to start Sentaurus Mesh and provides a

general explanation of its functionality.

Overview
Sentaurus Mesh is a suite of tools that produce finite-element meshes for use in applications
such as semiconductor device simulations, process simulations, and electromagnetic
simulations. It has three mesh generation engines: an axis-aligned mesh generator, an offsetting
mesh generator, and a tensor-product mesh generator. Sentaurus Mesh also provides a set of
tools that perform operations on boundary representations and meshes.

The axis-aligned and offsetting mesh generators produce Delaunay meshes, which are suitable
for use in Sentaurus Device and Sentaurus Process. In one dimension, the meshes contain
segments only. In two dimensions, the meshes contain triangles only, while in three
dimensions, the meshes comprise tetrahedra. For information about the algorithm used to
generate Delaunay meshes, see Chapter 6 on page 113.

The offsetting mesh generator can produce layered meshes in two and three dimensions. The
layers are located at the device interfaces and follow the contours of the interface. They can be
combined with axis-aligned elements to produce high-quality meshes for nonplanar structures.
As such, the offsetting mesh generator is a superset of the axis-aligned mesh generator, where
layering takes precedence over axis-aligned mesh generation.

The tensor-product mesh generator is intended to generate meshes for Sentaurus Device
Electromagnetic Wave Solver and for some applications in Sentaurus Device. The meshes
contain rectangular elements in two dimensions and cuboid elements in three dimensions.

Sentaurus Mesh reads the input geometry from a boundary file stored in the TDR format with
the _bnd.tdr file extension. Some TDR files from Sentaurus Process and Sentaurus
Interconnect with the _fps.tdr and _sis.tdr file extensions, respectively, contain two
geometry objects: one for the volumetric data and one for the boundary representation.
Sentaurus Mesh reads the boundary object in these TDR files, but it ignores other geometry
objects.

See Sentaurus™ Data Explorer User Guide, Appendix B on page 117 for details about the
TDR file structure.

Sentaurus™ Mesh User Guide 1

O-2018.06
1: Introduction to Sentaurus Mesh
Applications of Different Mesh Generators

Impurity concentrations and user-required element sizes can be described using a mesh
command file. The grid can be adapted to analytic profiles generated by Sentaurus Structure
Editor or profiles generated by Sentaurus Process. (All references to concentrations in this
document imply ‘active’ or ‘substitutional’ concentrations, since calculations in Sentaurus
Device use concentrations in this form.)

The required point density is obtained by refining the elements in an anisotropic way.
Therefore, unnecessary point propagation due to quadtrees, octrees, or tensor-product grid
techniques is avoided.

A delaunization process allows Sentaurus Mesh to obtain high-quality conforming Delaunay

grids, suitable for control volume discretization methods that are used in device simulation. For
more information, refer to the literature [1][2][3][4][5].

The output of Sentaurus Mesh depends on the mesh generation engine used. The axis-aligned
and offsetting mesh generators always produce a TDR unstructured mesh; the tensor-product
mesh generator will select the type of mesh depending on the target application.

Applications of Different Mesh Generators

The choice of which mesh generator to use for a particular application depends largely on the
geometry of the device.

For devices where the most important surfaces are axis aligned, the recommendation is to use
the axis-aligned mesh generator, since it produces the highest quality elements with minimal
node count for such devices.

For devices where the main surfaces are nonaxis-aligned or curved (for example, a MOS-type
structure where the channel is nonplanar), the recommendation is to use the offsetting mesh
generator, since it produce meshes containing layers that better conform to the curved surfaces,
thereby reducing the number of elements in the final mesh (see Offsetting Section on page 38).

For electromagnetic simulations using Sentaurus Device Electromagnetic Wave Solver, use the
tensor-product mesh generator.

2 Sentaurus™ Mesh User Guide

O-2018.06
 1: Introduction to Sentaurus Mesh
Starting Sentaurus Mesh

Starting Sentaurus Mesh

In Sentaurus Mesh, a mesh is created from two input files, namely, the boundary file and the
command file. If the input project is called project_name, a mesh can be created using the
command:
snmesh [<options>] project_name

Sentaurus Mesh automatically adds the extensions _bnd.tdr and .cmd to the base name
project_name. Sentaurus Mesh creates the output file project_name_msh.tdr that
contains mesh geometry information and doping information. Another file,
project_name_msh.log, is created and is used as the log file for the mesh generation.

Command-Line Options
The binary of Sentaurus Mesh is snmesh. It is executed using the syntax:
snmesh [<options>] <command_file_name>

Table 1 Command-line options available for Sentaurus Mesh

Option Description

-backcompat <release> Changes the behavior of the mesh generation and solid-modeling algorithms and
variable defaults to those of a specified previous release. For example:
-backcompat N-2017.09-SP1

-h Displays help information.

-v Displays version information only.

References
[1] L. Villablanca, Mesh Generation Algorithms for Three-Dimensional Semiconductor
Process Simulation, Series in Microelectronics, vol. 97, Konstanz, Germany: Hartung-
Gorre, 2000.
[2] P. Conti, M. Tomizawa, and A. Yoshii, “Generation of Oriented Three-Dimensional
Delaunay Grids Suitable for the Control Volume Integration Method,” International
Journal for Numerical Methods in Engineering, vol. 37, no. 19, pp. 3211–3227, 1994.
[3] G. Garretón et al., “A New Approach for 2-D Mesh Generation for Complex Device
Structures,” in International Workshop on Numerical Modeling of Processes and
Devices for Integrated Circuits (NUPAD V), Honolulu, HI, USA, pp. 159–162, June
1994.

Sentaurus™ Mesh User Guide 3

O-2018.06
1: Introduction to Sentaurus Mesh
References

[4] G. Garretón et al., “Unified Grid Generation and Adaptation for Device Simulation,” in
Simulation of Semiconductor Devices and Processes (SISDEP), vol. 6, Erlangen,
Germany, pp. 468–471, September 1995.
[5] G. Heiser, Design and Implementation of a Three-Dimensional General Purpose
Semiconductor Device Simulator, Series in Microelectronics, vol. 13, Konstanz,
Germany: Hartung-Gorre, 1991.

4 Sentaurus™ Mesh User Guide

O-2018.06
CHAPTER 2 Command File

This chapter describes the sections of the command file of Sentaurus

Mesh.

Overview
In the command file (.cmd), you can specify different parameters for the generation of a mesh
as follows:
■ Sections are delimited by opening and closing braces.
■ Only one keyword must be specified per line.
■ Keywords used in the command file are not case sensitive.
■ Strings are enclosed in double quotation marks.
■ Comments start with * or #.

Different types of information can be given in the command file. You can specify refinement
information, doping profile information, and control parameters for the different mesh
generators and tools provided in Sentaurus Mesh.

Refinement information is required to control mesh generation according to user requirements

(local element size). This information is specified in the Definitions section. Profile
information is required to define the fields, for example, doping profiles, which are used in grid
adaptation. Doping profiles can be specified with different types of information:
■ External simulation results
■ Constant data
■ Analytic formulas and predefined functions describing a profile

The command file has the following sections:

■ The command file can start with an optional title statement, which consists of the Title
keyword followed by a string in double quotation marks. By default, Title "" is used.
■ IOControls specifies an explicit input file containing the structure and an output file to
which the generated mesh will be saved.
■ Definitions defines the sets of refinement parameters and profile definitions to be used
in the Placements section. These sets are referred to using their unique reference name.
■ Placements defines instances of the definitions given in the Definitions section,
placed with respect to the current device.

Sentaurus™ Mesh User Guide 5

O-2018.06
2: Command File
IOControls Section

■ Interpolate controls data interpolation.

■ AxisAligned controls the axis-aligned mesh generator.
■ Offsetting controls the offsetting mesh generator.
■ Delaunizer controls the behavior of the delaunizer in Sentaurus Mesh.
■ Tensor controls the tensor-product mesh generator.
■ Tools specifies additional meshing utilities available in Sentaurus Mesh.
■ QualityReport specifies the mesh quality statistics to be reported and the limits for the
mesh quality criteria.

The syntax of the command file is:

Title ""
IOControls {input/output information}
Definitions {defining information}
Placements {placing information}
Interpolate {data interpolation information}
AxisAligned {axis alignment information}
Offsetting {offsetting information}
Delaunizer {delaunizer information}
Tensor {tensor information}
Tools {tools information}
QualityReport {mesh quality information}

The different sections of the command file of Sentaurus Mesh are described in the next
sections.

IOControls Section
The IOControls section is used to specify the names of input files describing the structure
and the name of the output file with the generated result. The input and boundary files can
contain either a boundary or a mesh in TDR format.

You can use the EnableOffset, EnableSections, EnableTensor, and EnableTools

options to enable different algorithms based on the contents of the command file. The result
after enabling unrelated sections in the command file is undefined.

The syntax of this section is:

IOControls {
EnableEMW
EnableOffset
EnableSections
EnableTensor

6 Sentaurus™ Mesh User Guide

O-2018.06
 2: Command File
IOControls Section

EnableTools
inputFile = "string"
numThreads = integer
outputFile = "string"
useDFISEcoordinates
useUCScoordinates
verbosity = 0 | 1 | 2 | 3
}

where (default values are given in parentheses if applicable):

EnableEMW
Generates meshes suitable for Sentaurus Device Electromagnetic Wave Solver (EMW)
applications using the tensor-product mesh generator (see EMW Subsection for
Computing Cell Size Automatically on page 49).

EnableOffset
Enables the Offsetting section of the command file and the offsetting mesh generator
(see Offsetting Section on page 38).

EnableSections
Parses the command file and activates the mesh generators associated with the sections
present in the command file. If the command file contains the AxisAligned, Tools,
Tensor, or Offsetting sections, the EnableSections option activates automatically
the corresponding mesh generators.

EnableTensor
Enables the tensor-product mesh generator (see Tensor Section on page 45).

EnableTools
Enables the operations that can be specified in the Tools section (see Tools Section on
page 54).

inputFile
The name of the default input file is based on the name of the command file. If an input file
is specified, it is used as the input file instead of the default input file based on the name of
the command file.

numThreads (1)
Sets the number of threads to be used by the mesh generators (axis-aligned, offsetting, and
tensor-product).

Sentaurus™ Mesh User Guide 7

O-2018.06
2: Command File
Definitions Section

outputFile
Specifies the name of the output file.

useDFISEcoordinates
Converts all coordinates to the DF–ISE coordinate system (except the coordinates from the
command file).

useUCScoordinates
Uses the unified coordinate system (UCS). With the exception of the command file, the
coordinates from all files read by Sentaurus Mesh are converted to the UCS.

verbosity
Sets the verbosity level of the output messages. At level 0, only basic messages are
displayed. At level 3, all messages are displayed.

Definitions Section
The Definitions section is composed of sets of refinement and profile subsections. Each
subsection consists of a reference name, an opening brace, the specification of parameters, and
a closing brace.

The order of definitions in the Definitions section is not important since these definitions
are used as references in the Placements section.

The syntax of this section is:

Definitions {
Refinement "reference name" {parameters}
Multibox "reference name" {parameters}
Constant "reference name" {parameters}
AnalyticalProfile "reference name" {parameters}
SubMesh "reference name" {parameters}
Particle "reference name" {parameters}
...
}

8 Sentaurus™ Mesh User Guide

O-2018.06
 2: Command File
Definitions Section

Defining Refinement Regions

The syntax to define a refinement region is:
Refinement "reference name" {
MaxElementSize = value | vector
MinElementSize = value | vector
RefineFunction = MaxGradient(parameters) | MaxTransDifference(parameters) |
MaxInterval(parameters) | MaxLengthInterface(parameters)
}

where (default values are given in parentheses if applicable):

MaxElementSize (1)
Controls the maximum size of the grid elements (you also can use its abbreviation
MaxElemSize). A real number or a vector x = [ x 1, …, x d ] can be specified, where d is
the dimension and x d represents the maximum edge lengths along the coordinate axes. A
vector can be used to refine nonisotropically. Only values greater than zero are considered.

MinElementSize (0.02)
Controls the minimum size of the grid elements (you also can use its abbreviation
MinElemSize). A real number or a vector x = [ x 1, …, x d ] can be specified, where x d
represents the minimum edge lengths along the coordinate axes. Grid elements can be
refined in one direction if their edge length in that direction is greater than the specified
value. Only values greater than zero are considered.

RefineFunction (MaxTransDifference)
Different functions can be used to select grid elements for refinement:
• MaxGradient (or use its abbreviation MaxGrad): The gradient of a profile
(Variable) in the element is evaluated. If the gradient is greater than Value and the
edge lengths are large enough, the element is refined. The syntax is:
RefineFunction = MaxGradient(Variable = "Dataset name",
Value = value | vector | tensor)
• MaxTransDifference (or use its abbreviation MaxTransDiff): The maximum
difference of the transformed values of a profile at the vertices of the element is
evaluated. If the difference is greater than Value and the edge lengths are large enough,
the element is refined. The syntax is:
RefineFunction = MaxTransDifference(Variable = "Dataset name",
Value = value | vector | tensor)

Sentaurus™ Mesh User Guide 9

O-2018.06
2: Command File
Definitions Section

The transformation applied to the values used in the refinement functions (linear,
logarithmic, arsinh) is defined in the datexcodes.txt file for each Variable (see
Utilities User Guide, Variables on page 2).
RefineFunction can be repeated for different variables in the same Refinement
section. If Variable is not defined, the default is "DopingConcentration". If
Value is not specified, it defaults to 1; however, no RefineFunction is assigned by
default.
Variable defines the dataset used to adapt the grid. The grid can be adapted according
to species or any type of variable defined in the output file. The values are computed
from the analytic formulas, constant data, and external simulation results defined in the
command file. Therefore, the name of a variable must match the name of a variable
stored in the output file. The variable name must be enclosed in double quotation
marks.
The parameter Value can be used to refine scalar, vector, or tensor variables.
To refine on a vector variable, use a vector of values, one per direction. For example,
in two dimensions, you can refine on ElectricField as follows:
RefineFunction = MaxTransDiff(Variable = "ElectricField",
Value = (100,100) )
To refine on a tensor variable, use an array of 9 elements where each component is
represented like this: (xx xy xz yx yy yz zx zy zz). Alternatively, if the tensor
field has symmetric components, you can use a 6-element array like this: {xx xy yy
yz zx zz}. For example, you can refine on Stress as follows:
RefineFunction = MaxTransDiff(Variable = "Stress",
Value = (1e10 2e10 1e10 1e8 1e10 1e10 2e10 1e10 1e8) )
The refinement is applied independently to each component of the vectors and tensors.
• MaxInterval: This function analyzes each edge in a refinement tree cell and refines
the edge if the data values at the endpoints overlap a given interval and the edge is
longer than the maximum edge length defined on the interval. The syntax is:
RefineFunction = MaxInterval(Variable = "Dataset name",
cmin = value | vector | tensor, cmax= value | vector | tensor,
targetLength = value, scaling = value, rolloff)
If the values at the edge endpoints overlap the range given by cmin and cmax, the
algorithm checks only whether the edge length is shorter than the targetLength
value. If this happens, the edge will be split.
When the edge is outside the value range, and the rolloff variable is true, the tool
adjusts targetLength to have a smooth transition into the coarser areas. To do this,
the tool applies the following formula:
targetLengthOutside = targetLength*(1 + log(Ca) – log(Cb))^2 * scaling

10 Sentaurus™ Mesh User Guide

O-2018.06
 2: Command File
Definitions Section

where Ca and Cb are the variable values at the endpoints of the edge.
• MaxLengthInterface (or use its abbreviation MaxLenInt): This function produces
refinement at the interfaces. The syntax is:
RefineFunction = MaxLengthInterface(Interface("Material1","Material2"),
Value = value, Factor = value, DoubleSide, UseRegionNames)
RefineFunction can be repeated for different interfaces in the same Refinement
section.
The material specified in the Interface statement must be a valid DATEX material.
The first material indicates the side of the interface on which the refinement is
performed. To apply the refinement to both sides of the interface, specify the
DoubleSide option.

By default, interfaces are defined by a pair of materials. However, if the option

UseRegionNames is used, the interface is interpreted as a regionwise specification.

The material "All" can be used to specify all interfaces of a given material and an
empty string can be used to specify outer interfaces. In addition, the second argument
in an interface specification can be a contact indicated by either the string "Contact"
or the name of the contact (if UseRegionNames is specified).
If Interface is not defined, no interface will be refined. If Value is not specified, it
defaults to 1. The Factor parameter must be a number greater than or equal to 1. If
Factor is not defined, it defaults to a huge number, so only one layer is produced.

Defining Multibox Regions

NOTE Using the Multibox subsection is no longer recommended. Instead,
use interface refinement with MaxLengthInterface (see Performing
Interface Refinement on page 85).

A multibox is a special refinement box that specifies a graded refinement along the x-, y-, or
z-direction. You can specify the required minimum and maximum element sizes, and an
additional refinement ratio in all directions. The created mesh is graded using the specified
ratios (also observing the minimum and maximum element sizes). The syntax to define a
multibox refinement region is:
Multibox "reference name" {
MaxElementSize = value | vector
MinElementSize = value | vector
Ratio = (ratio_width, ratio_height, ratio_depth)
}

Sentaurus™ Mesh User Guide 11

O-2018.06
2: Command File
Definitions Section

where:
■ MaxElementSize and MinElementSize are the same as described in Defining
Refinement Regions on page 9.
■ Ratio controls the grading of the element sizes:

• ratio_width is the grading factor in the x-direction.

• ratio_height is the grading factor in the y-direction.

• ratio_depth is the grading factor in the z-direction (3D only).

Defining Constant Profiles

The syntax to define a constant profile is:
Constant "reference name" {
Species = "string"
Value = value
}

where:

Species
Specifies the species or variables for the constant profile.

Value
Specifies the value of the constant profile.

Defining Analytic Profiles

Profiles can be defined using simple analytic expressions, which have two components. The
first component Function represents the values along a direction defined as the normal
direction of the ReferenceElement. This is the primary direction. These values are
smoothed along the direction perpendicular to the normal, or lateral direction, using the second
component LateralFunction.

These expressions can be of the following types:

■ Predefined functions: Gaussian and error function
■ One-dimensional external profile
■ Your own function (using the general function evaluator)

12 Sentaurus™ Mesh User Guide

O-2018.06
 2: Command File
Definitions Section

The formulas used for these analytic profiles are described in Appendix A on page 117.

The syntax to define an analytic profile is (instead of AnalyticalProfile, you can use its
abbreviation AnaProf):
AnalyticalProfile "reference name" {
Species = "string"
Function = Gauss(primary parameters) | Erf(primary parameters) |
subMesh1D(primary parameters) |
Eval(primary parameters) | General(primary parameters)
LateralFunction = Gauss(lateral parameters) | Erf(lateral parameters) |
Eval(lateral parameters)
}

where:

Species
Specifies the species for the analytic profile.

Function
Indicates the type of function and the parameters used along the primary direction, the
direction normal to the ReferenceElement.

LateralFunction
Defines the lateral component of the analytic profile (you also can use its abbreviation
LatFunc). A Gaussian function, an error function, or a general analytic function can be
specified using the following lateral parameters:
LateralFunction = Gauss(Factor = value)
LateralFunction = Gauss(StandardDeviation = value)
LateralFunction = Gauss(Length = value)
LatFunc = Erf(Factor = value)
LatFunc = Erf(Length = value)
LatFunc = Eval(init = "..." function = "...")

NOTE The default LateralFunction is the error function Erf.

If you use General to specify the analytic profile, there is no separate

LateralFunction since the definition of the analytic profile using
General includes both the primary and the lateral directions in its
formulation.

Sentaurus™ Mesh User Guide 13

O-2018.06
2: Command File
Definitions Section

Specifying a Gaussian Function

A Gaussian function can be specified with the following primary parameters:

Function = Gauss(PeakPosition = value, PeakValue = value,
StandardDeviation = value)
Function = Gauss(PeakPosition = value, Dose = value, StdDev = value)
Function = Gauss(PeakPosition = value, PeakValue = value, Length = value)
Function = Gauss(PeakPosition = value, Dose = value, Length = value)
Function = Gauss(PeakPosition = value, PeakValue = value,
ValueAtDepth = value, Depth = value)
Function = Gauss(PeakPos = value, Dose = value, ValAtDepth = value,
Depth = value)

By default, PeakPosition=0. There are no default values for the other parameters.

If Function=Gauss, Factor=0.8 in LateralFunction by default.

Some parameters have abbreviations (provided in parentheses) you can use, including:
StandardDeviation (StdDev), PeakPosition (PeakPos), PeakValue (PeakVal), and
ValueAtDepth (ValAtDepth).

Specifying an Error Function

An error function can be specified with the following primary parameters:
Function = Erf(SymmetryPosition = value, MaxValue = value, Length = value)
Function = Erf(SymmetryPosition = value, Dose = value, Length = value)
Function = Erf(SymPos = value, MaxVal = value, ValAtDepth = value,
Depth = value)
Function = Erf(SymPos = value, Dose = value, ValAtDepth = value,
Depth = value)

By default, SymmetryPosition=0.

If Function=Erf, Factor=0.8 in LateralFunction by default.

Some parameters have abbreviations (provided in parentheses) you can use, including:
SymmetryPosition (SymPos) and MaxValue (MaxVal).

Specifying a 1D External Profile

To specify a 1D external profile, the syntax is:

Function = subMesh1D(Datafile = "string", DataScale = value,
Scale = value, Range = line [(x1), (x2)])

14 Sentaurus™ Mesh User Guide

O-2018.06
 2: Command File
Definitions Section

The Datafile parameter specifies a file in XGRAPH format, which consists of a title
enclosed in double quotation marks and a list of "x y" values. More than one profile can be
included in Datafile.

The DataScale parameter scales the data values contained in the data file. Each input value
is multiplied by the DataScale factor. By default, DataScale=1.

The Scale parameter scales the coordinate values from the file. By default, Scale=1.

The optional Range parameter selects a range of values from the file. The keywords x1 and x2
must be given in the file coordinate system. Range is applied to all profiles inside the file. By
default, the entire data range is selected.

If Function=subMesh1D, StandardDeviation=0.8 in LateralFunction by default.

Using the General Function Evaluator

The general function evaluator can be used in either of two ways:

■ Using Eval: A user-specified analytic function in the primary direction (normal to the
reference window) and a separate decay function (Gaussian, error function, or a user-
defined function with Eval) in the lateral direction. The syntax is:
AnalyticalProfile "reference name" {
Function = Eval(init = "string", function = "string", value = value)
LateralFunction = Eval(init = "string", function = "string")
}
■ Using General: A user-defined function specified directly in device coordinates. There is
no concept of primary and lateral directions because the General function is specified
directly as a function of the x-, y-, and z-direction. The syntax is:
AnalyticalProfile "reference name" {
Function = General(init = "string", function = "string", value = value)
}

NOTE General does not require LateralFunction since the General

function is evaluated directly in all device coordinates.

Both the Eval and General functions use the same syntax for the primary parameters. The
difference is that General uses spatial coordinates and Eval uses coordinates that are
measured in the primary or lateral profile direction when used to define the primary or lateral
profile, respectively.

Sentaurus™ Mesh User Guide 15

O-2018.06
2: Command File
Definitions Section

The keywords init, function, and value correspond to the initialization formula, the
evaluation formula, and the default value (in the case of a failed formula evaluation at a data
point) (for the use of General functions, see Using Analytic Functions for Refinement I on
page 94):

init
Specifies a semicolon-separated list of assignments for variables that are used later, for
example, init = "a=2;b=4". This string is evaluated only once.

function
Specifies an expression that is evaluated for every query. The variable that replaces the
primary or lateral distance must be called x, for example:
function = "sin(x)"
function = "exp(4*x)*sin(x)"
In general, 1D, 2D, and 3D simple analytic functions can be specified here. The variables
x, y, and z can be used to refer to the respective x-, y-, and z-spatial coordinates.

value
Specifies the default return value if the evaluation fails. The default is 1.0e18 for the
primary direction and 1 when used as LateralFunction.

Note that:
■ All defined variables are global variables. This means that, if init="a=1" is defined in
one function, the same value will be used in all functions. Resetting the variable value in
another function command will have no effect.
■ You can freely mix Eval with Gauss, Erf, and subMesh1D functions.
■ The symbols "pi" and "e" can be used in the expressions.
■ The functions that can be used are:
"sin", "cos", "tan", "asin", "acos", "atan", "sinh", "cosh", "tanh", "exp",
"log", "log10", "sqrt", "floor", "ceil", "abs", "hypot", "deg", "rad"
■ Numeric exponential constants can be specified as either "2*10^18" or "2e18".
■ As an extension to the Eval function, the General function assesses device coordinates
directly, (x, y) and (x, y, z), and does not use primary and lateral distances. Any lateral
functions and reference geometries (in the Placements section) are ignored.

16 Sentaurus™ Mesh User Guide

O-2018.06
 2: Command File
Definitions Section

Defining Submeshes
External simulation results given on a mesh can be used to define profiles in the device. The
external mesh must have the same spatial dimension as the device. The datasets defined on the
external mesh are interpolated to the newly generated mesh. The external profiles are called
submeshes.

The syntax to define a submesh is:

SubMesh "reference name" {
Geofile = "string"
...
Fields = "string", "string", ...
}

where:

Geofile
Specifies the name of a file with an external mesh. The file must be in TDR format. The
dimension of the external mesh must be the same as the dimension of the device.

NOTE Sentaurus Mesh uses a simplified version of the submesh syntax where
only the Geofile parameter must be specified.

Fields
Specifies a list of fields to be extracted from the submesh. The other fields are ignored and
are not used in the calculations or written to the output file.

Defining Particle Profiles

Particle definitions can be used to define profiles associated with discrete dopant distributions
obtained from kinetic Monte Carlo (KMC) simulations using Sentaurus Process Kinetic Monte
Carlo (Sentaurus Process KMC). A continuous profile is obtained from the discrete dopant
distribution by associating a doping function with each discrete dopant (see Appendix B on
page 129).

The syntax to define a particle profile is:

Particle "reference name" {
AutoScreeningFactor
BoundaryExtension = value
Divisions = value
DopingAssignment = "CIC" | "NGP" | "Sano"

Sentaurus™ Mesh User Guide 17

O-2018.06
2: Command File
Definitions Section

Normalization
NumberOfThreads = integer
ParticleFile = "string"
ScreeningFactor = value
ScreeningScalingFactor = value
Species = "string"
}

where (default values are given in parentheses if applicable):

AutoScreeningFactor
If this option is specified, Sentaurus Mesh calculates automatically a screening factor for
1⁄3
each discrete dopant based on the local density of dopants using k c = 2N ( x 0, y 0, z 0 ) ,
where N ( x 0, y 0, z 0 ) is the density at the location of the discrete dopant.
Even when this option is specified, ScreeningFactor also must be specified because,
when calculating the local density, the integration box size (in micrometers) is determined
4
using ( 4.4934/ ScreeningFactor ) × 10 .

BoundaryExtension
This parameter applies to 2D structures only and is used to obtain continuous doping on a
2D structure from a 3D KMC TDR file containing particle information.
The value given in micrometers is a thickness that is used internally to create an imaginary
3D structure by extruding the input 2D structure. This 3D structure is used to compute
doping information, and this information is transferred to the 2D mesh.

Divisions (10)
This parameter applies to 2D structures only and is used in conjunction with the
BoundaryExtension parameter. For each mesh point in a 2D structure, a number of
points equal to a number of divisions, each separated by an equal amount, is created in the
z-direction. The amount of separation is obtained by dividing the boundary extension with
the number of divisions. The doping is computed on all of these points, and an average
doping is assigned for the corresponding 2D mesh point.

DopingAssignment ("Sano")
The basic refinement method is the Sano method, but this parameter allows you to choose
a method by which doping is assigned to a mesh immediately before saving the mesh:
• The cloud-in-cell ("CIC") method distributes the doping of a particle to the vertex
nodes of the element in which the particle is located.
• The nearest grid point ("NGP") method assigns the doping of a particle to the nearest
mesh node.

18 Sentaurus™ Mesh User Guide

O-2018.06
 2: Command File
Placements Section

• The "Sano" method uses a doping function described in Appendix B on page 129 to
distribute the doping of a particle to surrounding nodes.

Normalization
Specifying this option compensates for doping loss of dopants located near the boundary.

NumberOfThreads (1)
Parallelizes the local screening factor computation. Multithreading is recommended if the
simulation contains thousands of particles.

ParticleFile
Specifies the name of the KMC TDR file that contains the particle (discrete dopant)
information.

ScreeningFactor
This is the cut-off parameter, k c , for the doping function associated with each discrete
–1
dopant (see Appendix B on page 129). The ScreeningFactor (given in units of cm )
can be used as a fitting parameter; however, a value for it can be estimated from
1⁄3
k c = 2N , where N is the impurity concentration.

ScreeningScalingFactor
Controls the degree of smoothness of the profile. It is applied to the screening factor when
AutoScreeningFactor is specified.

Species
Specifies the name of an active impurity concentration to associate with this definition, for
example, ArsenicActiveConcentration and BoronActiveConcentration.
If Species is not specified, all active impurities that are found in the KMC particle file
will be associated with this definition.

Placements Section
The Placements section is composed of sets of refinement and profile instances. Their
positions in the device must be specified, and they must reference a definition given in the
Definitions section. In other words, each instance or subsection consists of the instance
name, an opening brace, the specification of parameters, and a closing brace.

The order of the refinement regions in this section is important. The mesh generators select
which refinement condition will be applied depending on the order of the refinement regions
described in the Placements section.

Sentaurus™ Mesh User Guide 19

O-2018.06
2: Command File
Placements Section

The syntax of this section is:

Placements {
Refinement "instance name" {parameters}
Multibox "instance name" {parameters}
Constant "instance name" {parameters}
AnalyticalProfile "instance name" {parameters}
SubMesh "instance name" {parameters}
Particle "instance name" {parameters}
...
}

NOTE The order of the profile instances in the Placements section is

important only when the Replace option is used.

Geometric Elements
To specify Placements sections, you must use geometric elements. These elements are
geometric objects used to select or locate data, and they are not part of the grid elements. The
coordinates of these objects are defined relative to the coordinates of the device.

The allowed geometric elements and the number of coordinate values that must be specified
depend on the dimension n of the device.

Let x = [ x 1, …, x d ] denote a point. The following geometric elements are defined:

Point ( x 1 )

Line ( x 1, x 2 )

Rectangle ( x 1, x 2 )

Polygon ( x 1, …, x m ), m > 2

Complex polygon ( lump 1 ( polygon 1 ( x 1, ..., x m ), ..., polygon p ( x 1, ..., x m ) )

lump 1 ( polygon 1 ( x 1, ..., x m ), ..., polygon p ( x 1, ..., x m ) ) ), m > 2

Cuboid ( x 1, x 2 )

Polyhedron { polygon 1 ( x 1, …, x m ), …, polygon p ( x 1, …, x m ) } , m > 2

Simple polygons are closed internally by adding the line segment between x 1 = [ x 1, …, x d ]
and x m = [ x 1, …, x d ] . Only simple closed polyhedra are allowed. All their faces must be
described.

20 Sentaurus™ Mesh User Guide

O-2018.06
 2: Command File
Placements Section

Complex polygons are composed of lumps. Each lump represents a separate subpolygon,
possibly containing holes. The first polygon inside a lump is the outer contour of the lump,
while the subsequent polygons represent holes inside the lump.

The following is an example of the use of the complexPolygon element. The example
represents two separate loops, the first of which has a hole inside:
AnalyticalProfile "buried n-channel" {
Reference = "buried n-channel"
ReferenceElement {
Element = complexPolygon [
lump [polygon [( 1.0 0.0 2.0 ) ( 2.0 0.0 2.0 ) ( 2.0 1.0 2.0 )
( 1.0 1.0 2.0 )]
polygon [( 1.3 0.3 2.0 ) ( 1.6 0.3 2.0 ) ( 1.6 0.6 2.0 )
( 1.3 0.6 2.0 )]
]
lump [polygon [( 0.0 1.5 2.0 ) ( 0.5 1.5 2.0 ) ( 0.5 2.0 2.0 )
( 0.0 2.0 2.0 )]
]
]
Direction = negative
}
}

NOTE All polygons defined inside a complexPolygon element must be

coplanar.

To describe a polyhedron with arbitrarily oriented faces, use polygons

instead of rectangles.

Table 2 lists the geometric elements that can be used to specify different kinds of window in
each dimension.

Table 2 Geometric elements for specifying windows

Function 1D 2D 3D

EvaluateWindow in Placements section for profiles Line Rectangle, Polygon Cuboid, Polyhedron

ReferenceElement in Placements section for analytic Point Line Rectangle, Polygon

profiles

RefineWindow in Placements section for refinements Line Rectangle, Polygon Cuboid, Polyhedron

Sentaurus™ Mesh User Guide 21

O-2018.06
2: Command File
Placements Section

In addition to the above-defined geometric elements, Table 3 lists other non-geometric

elements that can be used to specify windows in the command file.

Table 3 Non-geometric elements for specifying windows

Element Syntax

Material element material [<list of DATEX material names>]

Region element region [<list of region names>]

Composite element element {<list of geometric elements>}

Sweep element sweepElement {<sweep element parameters>}

Refinement or evaluation windows can be restricted to work on a particular material or region

using the keyword material or region. For material, the argument is a valid DATEX
material name in brackets. For region, the argument is a valid (existing) region name in
brackets (see Regionwise and Materialwise Refinement on page 77).

In addition, elements can be combined to build more complex elements called composite
elements, which are useful when defining complex reference elements for analytic profiles (see
Using Composite Elements on page 76).

Sweep elements can be used to create 3D profiles by sweeping 2D profiles in 3D space. There
are two types of sweep element: path sweep and angle sweep (see Creating 3D Profiles From
2D Cross Sections on page 79).

Placing Refinement Regions

In the Placements section, a refinement instance is specified by a name, an opening brace,
the specification of parameters, and a closing brace. Several refinement instances can refer to
the same set of refinement parameters.

The syntax for a refinement instance is:

Refinement "instance name" {
Reference = "string"
RefineWindow = geometric element | material [<list>] | region [<list>]
}

where:

Reference
Specifies the reference to a previously defined refinement.

22 Sentaurus™ Mesh User Guide

O-2018.06
 2: Command File
Placements Section

RefineWindow
Defines the location of the refinement instance in the device. (Instead of RefineWindow,
you can use its abbreviation RefineWin.) By default, RefineWindow is the bounding box
of the device. Table 2 on page 21 lists the geometric elements that can be used. In addition,
you can specify regionwise or materialwise refinement, or both refinements (see
Regionwise and Materialwise Refinement on page 77).
You can specify RefineWindow multiple times in a Refinement section. When more
than one RefineWindow is present, Sentaurus Mesh only refines the common sections of
the refinement windows. This can be used to restrict the refinement to the part of a
refinement box lying inside a region or material. For example:
Refinement "Refinement along current flow under the oxide" {
Reference = "Refinement along current flow only in Silicon"
RefineWin = cuboid [ ( 4.4 0 1 ) , ( 7.6 1.8 3.5 ) ]
RefineWin = material ["Silicon"]
}

NOTE If no RefineWindow is specified, the refinement instance is used as the

default region for the entire device.

Placing Multibox Regions

In the Placements section, a multibox instance is specified by the keyword Multibox,
followed by the name of the multibox window and an opening brace. After the specification of
parameters, a closing brace is placed. Several multibox instances can refer to the same set of
multibox parameters.

The syntax for a multibox instance is:

Multibox "instance name" {
Reference = "string"
RefineWindow = geometric element
}

where:

Reference
Specifies the reference to a previously defined multibox.

RefineWindow
Defines the location of the refinement instance in the device. Table 2 on page 21 lists the
geometric elements that can be used. By default, RefineWindow is the bounding box of
the device.

Sentaurus™ Mesh User Guide 23

O-2018.06
2: Command File
Placements Section

NOTE If RefineWindow is not specified, the refinement instance is used as

the default region for the entire device.

Placing Constant Profiles

The syntax for constant profiles is:
Constant "instance name" {
Reference = "string"
EvaluateWindow {
Element = geometric element | material [<list>] | region [<list>]
DecayLength = value | GaussDecayLength = value
}
LocalReplace
Replace
}

where:

Reference
Specifies the reference constant to use. Only references to constant profiles are allowed.

EvaluateWindow
Defines the domain where the profile is evaluated and a decay length is applied in the
vicinity of the window boundaries. The domain can be specified using a geometric element
(see Table 2 on page 21), as well as by referring to materials or regions. (Instead of
EvaluateWindow, you can use its abbreviation EvalWin.)
The decay function reduces round-off errors. It can be either an error function or a Gaussian
function. To use an error function, specify DecayLength (or you can use its abbreviation
DecayLen). For a Gaussian decay function, specify GaussDecayLength.
If EvaluateWindow is not defined, the transition between profiles is abrupt. If
DecayLength=0, no decay function is applied and the transition between
EvaluateWindow and its vicinity is abrupt. If DecayLength is negative, the profile is not
applied to points on the border of Element. By default, DecayLength=0 for all the
profiles.
For analytic, constant, and particle profiles, the default value of Element is the bounding
box of the device. For submeshes, the default value of Element is the bounding box of the
submesh. See the equations in Appendix A on page 117 for details.

24 Sentaurus™ Mesh User Guide

O-2018.06
 2: Command File
Placements Section

NOTE Avoid using EvaluateWindow when the profile is valid in the entire
device and no decay function is required. The evaluation of a geometric
element is time consuming.

The DecayLength and GaussDecayLength parameters do not apply

to particle profiles.

LocalReplace
With the Replace option, all the computed species are set to zero and are set with the value
corresponding to the given profile instance. With the LocalReplace option, only species
defined in the corresponding Definitions section are set exclusively to zero and are
recomputed using the current profile instance. The other species are not updated.
Accordingly, the net doping contribution is updated. By default, LocalReplace is
switched off.

Replace
In general, the values for each profile at each point of the newly generated mesh are
computed as the sum of all profile instances defined in the Placements section. The
instances are inspected in the same order as they are defined in the command file. If
Replace is specified for a given instance, all current summed values are replaced by the
value corresponding to the given profile instance. By default, Replace is switched off.

Placing Analytic Profiles

The syntax for an analytic profile is:
AnalyticalProfile "instance name" {
Reference = "string"
ReferenceElement {
Element = element
Direction = positive | negative
}
EvaluateWindow {
Element = geometric element | material [<list>] | region [<list>]
DecayLength = value | GaussDecayLength = value
}
LocalReplace
NotEvalLine
Replace
}

Sentaurus™ Mesh User Guide 25

O-2018.06
2: Command File
Placements Section

where:

Reference
Specifies the analytic profile to use. Only references to analytic profiles are allowed.

ReferenceElement
The direction of the normal to the ReferenceElement defines the direction of the
analytic profile (instead of ReferenceElement, you can use its abbreviation RefElem).
When evaluating the function values, the mesh points of the newly generated mesh are
projected to the Element. The distance in the normal direction is used to evaluate the
Function. The distance of the projection to the boundary of Element is used to compute
the LateralFunction. By default, values are computed on both sides of the Element.
If Direction is specified, function values are computed only on the positive or negative
side of the Element.
In 1D devices, Element is a point, and the positive and negative directions are given
by the coordinate axis. In 2D devices, Element is a line and the positive direction is taken
to the right of the line.
In 3D devices, Element can be either a rectangle or polygon. The normal for a rectangle
must be one of the coordinate axes. The positive and negative directions are defined
from this axis. A (planar) polygon can be arbitrarily oriented in three dimensions. The
direction is defined by the order of the points defining the polygon. A polygon is
considered correctly oriented if the side of the polygon, which is surrounded by points in a
positive orientation, defines the positive direction. There is no default value for Element
and Direction.

EvaluateWindow
Restricts the placement of the analytic profile to a particular window, material, or region.
See description in Placing Constant Profiles on page 24.

LocalReplace
See description in Placing Constant Profiles on page 24.

NotEvalLine
If this option is specified, the profile is not evaluated at the location of the reference
element. This can be useful for placing two identical analytic profiles back-to-back using
opposite directions, but without evaluating the reference element twice.

Replace
See description in Placing Constant Profiles on page 24.

26 Sentaurus™ Mesh User Guide

O-2018.06
 2: Command File
Placements Section

Placing Submeshes
The syntax for references to submeshes in the Placements section is:
SubMesh "instance name" {
Reference = "string"
Reflect = X | Y | Z
Rotation {
Angle = value
Axis = value
}
ShiftVector = vector
EvaluateWindow {
Element = geometric element | material [<list>] | region [<list>]
DecayLength = value | GaussDecayLength = value
}
Ignoremat
LocalReplace
MatchMaterialType
Replace
}

where:

Reference
Specifies the reference submesh to use. Only references to profiles that are defined as
SubMesh are allowed.

Reflect
Specifies a reflection perpendicular to the specified coordinate axis. The allowed axes
depend on the dimension of the device. The reflection point (or line or plane) is placed at
the specified coordinate axis.

Rotation
Performs a counterclockwise rotation around the axis. The center of the rotation is the
origin of the coordinate system. By default, Angle=0. By default, Axis=Z (for two and
three dimensions). In one dimension, Rotation is not supported.

NOTE The ShiftVector, Reflect, and Rotation operations are

performed in the order they appear in the command file. The final
location and orientation of the submesh depends on this order.

Sentaurus™ Mesh User Guide 27

O-2018.06
2: Command File
Placements Section

ShiftVector
Translates a submesh to a new location. The vector is specified as two or three coordinates
enclosed by parentheses.

EvaluateWindow
Restricts the placement of the submesh to a particular window, material, or region. See
description in Placing Constant Profiles on page 24.

Ignoremat
If this option is specified, the material in submeshes is ignored. The standard behavior of
submesh interpolation is that the interpolated value is only accepted if the point is in a
region with the same material.
The option Ignoremat allows Sentaurus Mesh to always accept the interpolation. (By
default, if the materials do not match, the closest region with the correct material is
searched.) The default behavior is not checked. For example:
Placements {
SubMesh "NoName_0" {
Reference = "NoName_0"
Ignoremat
}
}

LocalReplace
See description in Placing Constant Profiles on page 24.

MatchMaterialType
When this option is specified, the submesh attempts to match equivalent material types (for
example, semiconductor, insulator, conductor) instead of trying to match material names
when looking up values from which to interpolate.

Replace
See description in Placing Constant Profiles on page 24.

Placing Particle Profiles

The syntax for placing particle profiles in the Placements section is:
Particle "instance name" {
Reference = "string"
EvaluateWindow {
Element = material [<list>] | region [<list>]

28 Sentaurus™ Mesh User Guide

O-2018.06
 2: Command File
Interpolate Section

}
LocalReplace
Replace
}

where:

Reference
Specifies the reference particle to use. Only references to particle profiles are allowed.

EvaluateWindow
See description in Placing Constant Profiles on page 24.

LocalReplace
See description in Placing Constant Profiles on page 24.

Replace
See description in Placing Constant Profiles on page 24.

Interpolate Section
The optional Interpolate section controls data interpolation that is performed after the mesh
generators have finished.

The syntax of this section is:

Interpolate {
interpolateElements = true | false
keepTotalConcentration = true | false
lateralDiffusion = true | false
}

where (default values are given in parentheses if applicable):

interpolateElements (false)
Interpolates element-type (scalar and vector) datasets. The element-type datasets of the
input submesh are interpolated on the generated grid. The default value ignores element-
type datasets.

Sentaurus™ Mesh User Guide 29

O-2018.06
2: Command File
AxisAligned Section

keepTotalConcentration (false)
Saves the TotalConcentration field in the output file. By default, Sentaurus Mesh does
not save this field in the output file. Sentaurus Device can calculate this field, if necessary,
based on the available dopants.

lateralDiffusion (false)
Enables lateral extension on analytic profiles like that performed by the Synopsys Taurus™
Medici tool. This parameter affects only profiles with rectangular reference elements and
attenuates the lateral decay factor by taking into account the distance from the interpolated
points to all sides of the rectangle (see Lateral Error Function on page 126).

AxisAligned Section
The AxisAligned section controls the axis-aligned mesh generator in Sentaurus Mesh.

The axis-aligned mesh generator takes a boundary representation of the device and a series of
user-defined refinement criteria, and follows these steps to create a mesh:
1. It attempts to repair the boundary using a combination of decimation and reconstruction
algorithms. The algorithms are controlled by the geometricAccuracy parameter as well
as parameters related to the Delaunay refinement for piecewise smooth complex (DelPSC)
algorithm.
2. It produces an initial coarse discretization of the bounding box of the structure by applying
the xCuts, yCuts, and zCuts parameters. This creates an initial tensor-like structure that
is used as the basis for the user-defined refinement.
3. The basic mesh is refined using user-defined criteria described in Chapter 3 on page 73.
Each box is bisected recursively until all resulting boxes meet the criteria specified by the
user. During this process, the mesh generator ensures that criteria such as maxAngle,
maxAspectRatio, and maxNeighborRatio are satisfied.

4. After the boxes have been refined, they are imprinted on the boundary, producing a surface
axis-aligned pattern. At this stage, short surface edges and poor angles are eliminated using
a combination of boundary decimation and boundary repair algorithms.
5. As the last step before delaunization, the boxes are merged with the boundary, ensuring that
intersecting the boxes with the boundary does not produce an unbalanced mesh (that is, a
short edge next to a long one). To control this, the algorithm uses the parameter
maxBoundaryCutRatio.

30 Sentaurus™ Mesh User Guide

O-2018.06
 2: Command File
AxisAligned Section

The syntax of the AxisAligned section is:

AxisAligned {
allowRegionMismatch = true | false
binaryTreeSplitBox = (floatlist)
binaryTreeSplitFactorX = integer
binaryTreeSplitFactorY = integer
binaryTreeSplitFactorZ = integer
convexTriangulation = true | false
coplanarityAngle = float
decimate = true | false
DelPSC = true | false
DelPSCAccuracy = float
DelPSCRidgeAngle = float
DelPSCRidgeSampling = float
DualContouring = true | false
DualContouringDecimation = true | false
DualContouringMinAngle = float
DualContouringMinDihedralAngle = float
DualContouringResolution = float
fitInterfaces = true | false
geometricAccuracy = float
hintBoxSize = float
imprintAccuracy = float
imprintCoplanarFacesOnly = true | false
imprintCoplanarityAngle = float
imprintCoplanarityDistance = float
latticeCellSize = (float float float)
latticeDimensions = (integer integer integer)
maxAngle = float
maxAspectRatio = float
maxBoundaryCutRatio = float
maxNeighborRatio = float
minEdgeLength = float
minimumRegionMismatchVolume = float
overscan = true | false
overscanResolution = float
skipSameMaterialInterfaces = true | false
smoothing = true | false
spacingMethod = even | regular | smooth
splitDisconnectedRegions = true | false
virtualSpacing = true | false
xCuts = (floatlist)
yCuts = (floatlist)
zCuts = (floatlist)
}

Sentaurus™ Mesh User Guide 31

O-2018.06
2: Command File
AxisAligned Section

where (default values are given in parentheses if applicable):

allowRegionMismatch (false)
If allowRegionMismatch = true, when Sentaurus Mesh checks whether the number
of regions in the input boundary and the number of regions at the end of the meshing
process are the same, if there is a difference between the numbers of regions, Sentaurus
Mesh will ignore the discrepancy, and the meshing process will continue.
If allowRegionMismatch = true and minimumRegionMismatchVolume has also
been specified, Sentaurus Mesh checks the volumes of all deleted regions:
• If the volume of a deleted region is less than the value specified by
minimumRegionMismatchVolume, the meshing process will continue and the
number of deleted regions is reported.
• If the volume of a deleted region is greater than the value specified by
minimumRegionMismatchVolume, the meshing process will stop.
If allowRegionMismatch = false, when Sentaurus Mesh checks whether the number
of regions in the input boundary and the number of regions at the end of the meshing
process are the same, if there is a difference between the numbers of regions, the meshing
process will stop.

binaryTreeSplitBox
Specifies a box that defines the region where binaryTreeSplitFactorX,
binaryTreeSplitFactorY, and binaryTreeSplitFactorZ are applied. By default,
no box is used.
For 2D simulations, binaryTreeSplitBox is set to:
(xmin <float> ymin <float> xmax <float> ymax <float>)
For 3D simulations, binaryTreeSplitBox is set to:
(xmin <float> ymin <float> zmin <float> xmax <float> ymax <float> zmax <float>)

binaryTreeSplitFactorX (1)
Instructs Sentaurus Mesh to split the final binary tree used in the refinement step by a
specified factor in the x-direction. This factor must be a power of 2; otherwise, the nearest
power of 2 will be used. This parameter can be used to achieve an approximately uniform
mesh refinement in the x-direction.

binaryTreeSplitFactorY (1)
Same as binaryTreeSplitFactorX but in the y-direction.

32 Sentaurus™ Mesh User Guide

O-2018.06
 2: Command File
AxisAligned Section

binaryTreeSplitFactorZ (1)
Same as binaryTreeSplitFactorX but in the z-direction.

convexTriangulation (false)
Creates a minimum triangulation of a 3D model containing convex regions. The input 3D
boundary must be a convex model. Since the goal is to create a minimum triangulation of
the convex model, the refinement specifications in the command file (if any) are ignored.
If the input model contains nonconvex regions, the meshing terminates with a
corresponding message.

coplanarityAngle (175)
Specifies the angle used during boundary decimation to determine whether two faces are
coplanar.

decimate (true)
Specifies whether the 3D boundary is decimated. The decimation process removes nodes
from the surface, thereby generating a simpler structure. A node is removed only if the
deformation caused by removing the node is less than the value specified by the
geometricAccuracy parameter.

DelPSC (false)
Instructs Sentaurus Mesh to apply the DelPSC algorithm to boundary surfaces. It is useful
for curved surfaces. If you specify IOControls{numThreads=integer}, the DelPSC
algorithm uses multithreading. See [1][2] for a description of the algorithm.

DelPSCAccuracy (0.0001)
Controls the deviation (given in μm ) between the new curved surface and the original
curved surface in the DelPSC algorithm. The new curved surface can deviate from the
original curved surface by, at most, the value of DelPSCAccuracy. New vertices lie
exactly on the original surface, but new triangles cannot lie exactly on the original surface
unless the original surface is flat. In general, the smaller the value of DelPSCAccuracy is,
the smoother the new surface becomes, and the more accurate the new surface represents
the original surface.
In general, setting DelPSCAccuracy to 2% of the radius of curvature is appropriate.
Larger values allow the DelPSC algorithm to run faster but generate coarser discretization
on curved surfaces. Smaller values make the DelPSC algorithm run slower and generate
finer discretization on curved surfaces.

DelPSCRidgeAngle (150)
Angle used by the DelPSC algorithm to determine geometric features.

Sentaurus™ Mesh User Guide 33

O-2018.06
2: Command File
AxisAligned Section

DelPSCRidgeSampling (0.01)
Controls the size (given in μm ) of small triangles on curved surfaces in the DelPSC
algorithm.
In general, setting DelPSCRidgeSampling to 10% of the radius of curvature is
appropriate. Larger values allow the DelPSC algorithm to run faster but generate bigger
triangles next to geometric features and triple lines on curved surfaces. Smaller values
make the DelPSC algorithm run slower and generate smaller triangles next to geometric
features and triple lines on curved surfaces.

DualContouring (false)
Instructs Sentaurus Mesh to use the dual-contouring algorithm to reconstruct the boundary.
Dual contouring can be useful when the input surface is low quality, that is, it contains
triangles with small planar or dihedral angles. Dual contouring preserves sharp edges as
much as possible and avoids generating small features that cannot be meshed. The high-
quality reconstructed surface can then be passed to the mesher for volume mesh generation.
See [3][4][5] for a description of the algorithm.

DualContouringDecimation (false)
Instructs Sentaurus Mesh to apply a decimation algorithm to reduce the number of nodes
produced by the dual-contouring algorithm.

DualContouringMinAngle (5)
Determines the minimum planar angle (in degrees) on the surface produced by the dual-
contouring algorithm.

DualContouringMinDihedralAngle (10)
Determines the minimum dihedral angle (in degrees) on the surface produced by the dual-
contouring algorithm.

DualContouringResolution (0.002)
Specifies the minimum spacing (given in μm ) of an octree grid used for the surface
generation. The higher the resolution (smaller spacing), the more accurate the boundary
reconstruction. A very high resolution can result in an excessive number of triangles and a
longer runtime.

fitInterfaces (false)
Instructs Sentaurus Mesh to calculate the xCuts, yCuts, and zCuts automatically by first
refining along the axis-aligned interfaces.

34 Sentaurus™ Mesh User Guide

O-2018.06
 2: Command File
AxisAligned Section

geometricAccuracy (1e-6)
Restricts the changes to the boundary, which are undertaken by the decimation algorithm.
The decimation algorithm is not allowed to modify the boundary more than the value of
geometricAccuracy given in μm .

hintBoxSize (1.0)
When overscanning analytic profiles, Sentaurus Mesh calculates a hint box containing the
peak value. The size of this box is a number of standard deviations from the peak. The
default is one standard deviation around the peak value.

imprintAccuracy (1e-5)
Distance used to determine whether two points are too close during axis-aligned
imprinting.

imprintCoplanarFacesOnly (true)
If this option is switched on, Sentaurus Mesh imprints the axis-aligned refinement only on
faces that are away from curved regions of the boundary. This is useful to avoid
overrefinement in curved areas.

imprintCoplanarityAngle (179.9)
Angle used by the face-imprinting algorithm to determine whether two boundary faces are
coplanar.

imprintCoplanarityDistance (1e-5)
Distance used by the face-imprinting algorithm to determine whether two faces are
coplanar.

latticeCellSize
Specifies a tensor-like tessellation of the structure with a spacing that is as close as possible
to the defined cell size. If the virtualSpacing parameter is used, the generated lines
guide the refinement algorithm, but not all of these lines will necessarily be present in the
final mesh.

latticeDimensions
Specifies a tensor-like tessellation of the structure with the defined number of lines along
each direction. If the virtualSpacing parameter is used, the generated lines guide the
refinement algorithm, but not all of these lines will necessarily be present in the final mesh.

Sentaurus™ Mesh User Guide 35

O-2018.06
2: Command File
AxisAligned Section

maxAngle (90 in 2D, 165 in 3D)

Determines the maximum angle produced in the binary tree. In two dimensions, the default
is 90° .

maxAspectRatio (1e6)
Specifies the maximum aspect ratio allowed in the elements of the binary tree at the end of
the refinement step.

maxBoundaryCutRatio (0.01)
Defines the maximum-allowed ratio between adjacent segments on an axis-aligned box
intersecting the boundary. When a segment belonging to an axis-aligned box intersects the
boundary and the resulting cuts have a higher ratio than specified by this parameter, all
axis-aligned faces associated with this segment are disabled and are not allowed in the final
mesh.
For example, in the following figure, a candidate axis-aligned segment AC can intersect a
boundary edge at point B such that the AC segment becomes two colinear segments AB
and BC.
A

In the absence of maxBoundaryCutRatio (value 0), the ratio of the lengths of these edges
would be unconstrained. With this parameter, the ratio between the short segment and the
long segment are determined: either AB/AC or BC/AC. If either of these lengths is less than
maxBoundaryCutRatio, the AC segment and any other segment connected to it are
rejected and do not appear in the final mesh. In the case of a 3D mesh, this means that any
axis-aligned face touching this segment is disabled and cannot appear in the final mesh.
Setting maxBoundaryCutRatio to a high value (closer to 1) reduces the possibility of
having sharp changes in mesh sizes across the boundary. However, at the same time, it may
create holes in the mesh since, potentially, many axis-aligned faces or segments may be
disabled. Setting maxBoundaryCutRatio to a small value reduces the possibility of holes
around the boundary of the device, but it will produce sharp transitions in the mesh size at
the boundary.

maxNeighborRatio (2 in 2D, 4 in 3D)

Specifies the size ratio between adjacent elements.

36 Sentaurus™ Mesh User Guide

O-2018.06
 2: Command File
AxisAligned Section

minEdgeLength (1e-7)
Specifies the minimum edge length (given in μm ) produced on the boundary before the
delaunization step.

minimumRegionMismatchVolume (0)
Specifies a region volume that Sentaurus Mesh uses when checking deleted regions. It is
used in conjunction with allowRegionMismatch=true.

overscan (false)
Instructs Sentaurus Mesh to scan the axis-aligned cells for field changes that justify more
refinement based on the user parameters. The algorithm creates a small tensor mesh with
the resolution indicated by the overscanResolution parameter and uses it to check for
fine variations in the field profiles.
To avoid scanning all cells, the tool obtains ‘hints’ from the profiles as to where the
interesting areas are located. For this purpose, the tool internally calculates the location of
the peak values and p-n junctions, and gives them as a hint to the mesh generator.

overscanResolution (0.3)
Resolution used to scan the device for field changes. The algorithm takes each unrefined
cell and virtually subdivides it into smaller cells. Then, these small cells are checked for
changes in the field values that justify more refinement.

skipSameMaterialInterfaces (false)
During refinement, if this parameter is set to true, Sentaurus Mesh ignores interfaces that
have the same material on both sides.

smoothing (true)
Specifies whether the binary tree will be graded using the maxAspectRatio and
maxNeighborRatio parameters.

spacingMethod (even)
Specifies the type of progression used by the refinement algorithm when expanding the
refinement specified between lines:
• even: Distributes the cuts evenly, trying to approximate the spacing specified at the
beginning of the interval.
• regular: Distributes the cuts evenly using the exact spacing specified at the beginning
of the interval, and leaving the last interval with an approximate size if there is no more
room to accommodate the requested size.
• smooth: Distributes the cuts to have a smooth grading of spacing between lines.

Sentaurus™ Mesh User Guide 37

O-2018.06
2: Command File
Offsetting Section

splitDisconnectedRegions (false)
If an input boundary contains regions with multiple disconnected parts, this parameter
specifies whether these regions should be split into multiple disconnected regions and
renamed according to Sentaurus Process naming rules, or whether these regions and their
names should be preserved.

virtualSpacing (false)
Specifies whether the expansion lines produced by pairs of values defined by the xCuts,
yCuts, and zCuts parameters will be either explicit lines or virtual lines that guide the
refinement algorithm. When the lines are virtual, the refinement algorithm snaps the
refinement coordinates to these lines instead of using the standard bisection algorithm. This
allows the refinement to conform to a more user-defined pattern.

xCuts, yCuts, zCuts

These values represent refinement lines that are introduced into the mesh before any user-
defined refinement. The lines define a rectilinear grid from which to start the refinement.
Since each box in the initial grid is refined independently, different regions of the device
can be isolated, thereby obtaining a more predictable refinement in each one of them. The
cuts in each direction are specified as a list of cut points enclosed in parentheses.
Each cut point can be either a single floating-point value or a pair of floating-point values.
A single value indicates the position of the cut point. When a pair of values is used, the first
value indicates the position of the cut point, and the second value indicates the expansion
of the cuts into a sequence of lines to be generated between adjacent pairs of lines.

The following example creates a series of grid lines located at 0, 1, and 2 μm . Between 0 and
1 μm , the series of lines should have a spacing of 0.1 (10 lines). Between 1 and 2 μm , the
series of lines should have a spacing of 0.2 (5 lines):
xCuts = ( (0 0.1) (1 0.2) 2)
spacingMethod = even
The spacingMethod and virtualSpacing parameters control the refinement
algorithm.
By default, no cuts are introduced into the mesh unless the xCuts, yCuts, or zCuts
parameter is used.

Offsetting Section
The offsetting mesh generator uses the Offsetting section to create meshes with layers that
follow the device interfaces. The layers are combined with the axis-aligned mesh generated by
the axis-aligned mesh generator (see AxisAligned Section on page 30). The offsetting mesh

38 Sentaurus™ Mesh User Guide

O-2018.06
 2: Command File
Offsetting Section

generator first produces an axis-aligned mesh and then adds the offsetting layers on top of that
mesh, clearing the axis-aligned elements that overlap the layers.

NOTE Specify either the EnableOffset or EnableSections option in the

IOControls section of the command file to enable the offsetting mesh
generator (see IOControls Section on page 6).

The syntax of the Offsetting section is:

Offsetting {

# offsetting-global-section:
noffset {
factor = float
hlocal = float
maxlevel = integer
}

# offsetting-interface-section:
noffset material | region "string" "string" {
factor = float
hlocal = float
window = [(float float float) (float float float)]
}
# offsetting-region-section:
noffset material | region "string" {
maxlevel = integer
}
}

where (default values are given in parentheses if applicable):

factor (1.3)
As the front progresses, the thickness of the layers increases by this factor.

hlocal (0)
Thickness of first layer in μm . The default hlocal=0 means no layering at all.

maxlevel (200)
Specifies the number of layers that offsetting creates. If the front collides with other fronts
or surfaces, the mesh generator stops prematurely.

window
Specifies the cuboid used to confine the creation of layering. For a large interface, this
parameter allows you to limit the layering to a spatial region of interest, thereby reducing

Sentaurus™ Mesh User Guide 39

O-2018.06
2: Command File
Delaunizer Section

the size of the grid. Note that window controls only the start of the layering and, therefore,
the layers may grow outside of the window. Multiple windows can be specified within the
offsetting-interface section.

The parameters hlocal and factor affect the material interfaces and surfaces. They can be
specified on an interface basis using the syntax with two region names. The syntax is not
symmetric:
■ noffset region "A" "B" {} applies to the layers in region A where it borders region B.
■ noffset region "B" "A" {} sets parameters for the other side of the same interface.

At places where contacts are defined, their names are used to define interfaces. The
pseudo–region name Exterior is used for surfaces.

The parameter maxlevel can be set only per region using the syntax with one region name.

NOTE In both cases, you can use region names (keyword region) or the
material property (keyword material). Since the material property is
more persistent, it is advisable to use it instead of region names.

It is recommended to specify refinement criteria in the Definitions

and Placements sections of the command file. Otherwise, the
resulting mesh will be very coarse.

It is recommended to use a small number of layers to reduce spurious

refinements near the curved interfaces during delaunization of the mesh.

Delaunizer Section
The Delaunizer section controls the behavior of the delaunization algorithms in Sentaurus
Mesh. The syntax of this section is:
Delaunizer {
coplanarityAngle = float
coplanarityDistance = float
delaunayTolerance = float
edgeProximity = float
faceProximity = float
maxAngle = float
maxConnectivity = float
maxNeighborRatio = float
maxPoints = integer
maxSolidAngle = float
maxTetQuality = float
minAngle = float

40 Sentaurus™ Mesh User Guide

O-2018.06
 2: Command File
Delaunizer Section

minDihedralAngleAllowed = float
minEdgeLength = float
minEdgeLengthAllowed = float
sliverAngle = float
sliverDistance = float
sliverRemovalAlgorithm = integer
storeDelaunayWeight = true | false
type = boxmethod | conforming | constrained
}

where (default values are given in parentheses if applicable):

coplanarityAngle (175)
Determines whether two adjacent boundary faces are coplanar. The floating-point number
represents the angle between the faces.

coplanarityDistance (1e-5)
Determines whether two adjacent boundary faces are coplanar. The floating-point number
(given in μm ) represents the absolute deformation made to the surface when the common
edge is flipped.

delaunayTolerance (1e-4)
Specifies how close the ridges and boundary faces conform to the Delaunay criterion. A
value of 0 everywhere implies a very strict Delaunay criterion. A value of 1 everywhere is
equivalent to the construction of a constrained Delaunay triangulation (CDT). See
Delaunay Tolerance on page 44.

edgeProximity (0.05)
Specifies the minimum ratio of the length of a new edge to the length of the parent edge
from which it was generated. If an edge AB will be refined at point C and one of the ratios
AC/AB or CB/AB is smaller than edgeProximity, point C is moved to the center of AB.
When this value approaches 0.5, the edges will be more isotropically refined and the final
mesh may contain many more points.

faceProximity (0.05)
Specifies the minimum ratio of the area of a new face to the area of the parent face from
which it was generated. If a face ABC will be refined at point D and one of the ratios AD/
r, BD/r, or CD/r is smaller than edgeProximity (where r is the radius of the
circumscribed sphere), point D is moved to the Voronoï center of ABC. When this value
approaches 0.5, the faces will be more isotropically refined and the final mesh may contain
many more points.

Sentaurus™ Mesh User Guide 41

O-2018.06
2: Command File
Delaunizer Section

maxAngle (180)
Specifies the maximum angle allowed in the elements of the mesh (2D only).

maxConnectivity (1000)
Specifies the number of edges that can be connected to a mesh point.

maxNeighborRatio (1e+30)
Specifies the maximum-allowed ratio between the circumscribed spheres of neighboring
elements. Values close to 2 should give a better grading, but they may also increase the
mesh size considerably.

maxPoints (500000)
Sets a limit on the maximum number of points that the delaunizer generates. The limit is
observed after the ridges have been recovered.

maxSolidAngle (360)
Specifies the maximum solid angle allowed in the elements of the mesh (3D only).

maxTetQuality (1e37)
Specifies the maximum circumscribed sphere radius–to–shortest edge ratio allowed in the
mesh (3D only).

minAngle (360)
Specifies the minimum angle allowed in the elements of the mesh (2D only).

minDihedralAngleAllowed (0)
Checks that the 3D input boundary has a minimum dihedral angle greater than the value of
minDihedralAngleAllowed. Otherwise, Sentaurus Mesh terminates with an error
message (for 3D structures only).

minEdgeLength (1e-9)
A floating-point number (given in μm ) used to display a warning when the surface edges
become too short.

minEdgeLengthAllowed (0)
Checks that the 3D input boundary has a minimum edge length greater than the value of
minEdgeLengthAllowed. Otherwise, Sentaurus Mesh terminates with an error message
(for 3D structures only).

42 Sentaurus™ Mesh User Guide

O-2018.06
 2: Command File
Delaunizer Section

sliverAngle (175)
Controls the elimination of slivers. The sliver elimination algorithm removes all elements
where the maximum dihedral angle exceeds this value (given in degrees). The algorithm
endeavors to achieve this goal but, in general, it may not be possible. In practice, the final
meshes contain elements where the maximum dihedral angle is approximately 179° .

sliverDistance (1e-2)
Controls the amount of damage performed by the sliver elimination algorithm (see
Eliminating Slivers on page 115). The value specifies the maximum weight used at a given
node.

sliverRemovalAlgorithm (2)
Selects the sliver elimination algorithm:
• 1 selects the original algorithm.

• 2 selects the new algorithm that reduces the number of non-Delaunay elements by
assigning more appropriate weights to vertices (see Eliminating Slivers on page 115).

storeDelaunayWeight (true)
Stores the nodal weight from the sliver elimination algorithm in the output file as a field
variable when set to true. By supplying the Delaunay weight to Sentaurus Device, the box
method library will have better convergence. This field variable is called the
2
Delaunay–Voronoï weight (DelVorWeight) with the unit of μm in the TDR file.

type (boxmethod)
Specifies the type of Delaunay mesh that the delaunization algorithm constructs:
• The boxmethod option imposes very strict conditions on the boundaries. The smallest
circumscribed sphere around the boundary faces and ridges must be free of points.
• With the conforming option, the conditions at the boundary are more relaxed. This
means that there exists a circumscribed sphere around a boundary face, which is free
of points. This is equivalent to the standard Delaunay condition.
• When the constrained option is specified, the boundary faces are inserted into a
Delaunay mesh of the input points using a CDT algorithm. This option produces the
least refinement of all options, but it produces meshes that are not suitable for device
simulation.

Sentaurus™ Mesh User Guide 43

O-2018.06
2: Command File
Delaunizer Section

Delaunay Tolerance
The tolerance used to calculate the Delaunay criterion can be adjusted locally based on
region, material, or window information:
boundary material | region "string" "string" {
delaunayTolerance = float
window = { (float, float, float) (float, float, float) }
}
surface material | region "string" {
delaunayTolerance = float <WINDOW>
}
interior material | region "string" {
delaunayTolerance = float <WINDOW>
}

The delaunayTolerance parameter in the boundary, surface, and interior

subsections must always be specified. The window parameter is optional. These subsections
do not accept any other parameters (that is, you cannot restrict the values of parameters such
as maxPoints and minEdgeLength in regions or materials individually).

The following examples show the use of the Delaunay tolerance parameters:
Delaunizer {
# relax the tolerance at the boundary between any two materials
boundary {
delaunayTolerance=1
}

# restrict the tolerance at the boundary between silicon and oxide

boundary material "Oxide" "Silicon" {
delaunayTolerance = 1e-4
}

# relax the tolerance in the interior of the device

interior {
delaunayTolerance = 1
}

# restrict the tolerance around the gate area

interior region "gate" {
delaunayTolerance = 1
window = {(0.1,0,0) (0.2,0.1,0.1)}
}
}

44 Sentaurus™ Mesh User Guide

O-2018.06
 2: Command File
Tensor Section

Tensor Section
The Tensor section can contain the following subsections and controls the tensor-product
mesh generator:
Tensor {
Mesh {parameters}
EMW {parameters}
Box {parameters}
}

NOTE To activate the Tensor section, specify the EnableTensor option in

the IOControls section of the command file (see IOControls Section
on page 6).

Mesh Subsection for Controlling Mesh Generation

Various parameters can be defined in the Mesh subsection of a Tensor section of the command
file. These parameters control mesh generation. The syntax of a Mesh subsection is:
Tensor {
Mesh {
axisAlignedFeatureAngle = float
doping
grading = {float float float}
grading off
maxCellSize = float
minCellSize = float
maxBndCellSize = float
minBndCellSize = float
minNumberOfCells = integer
numPoints = integer
numPointsX = integer
numPointsY = integer
numPointsZ = integer
scale = {float float float}
window "string" float float float float float float
xCuts = (floatlist)
yCuts = (floatlist)
zCuts = (floatlist)
}
}

Sentaurus™ Mesh User Guide 45

O-2018.06
2: Command File
Tensor Section

where (default values are given in parentheses if applicable):

axisAlignedFeatureAngle (0.5 degrees)

Part of the process of generating the refinement involves refining the grid at the location of
axis-aligned interfaces found on the boundary. If there are boundary faces that are not
nearly axis-aligned, the refinement algorithm ignores them, leading to unexpected holes in
the refinement.
The axisAlignedFeatureAngle parameter allows you to specify a tolerance to indicate
to the tool which faces should be considered axis aligned. The tool measures the deviation
between the face normal and the nearest coordinate axis. If the deviation is smaller than
axisAlignedFeatureAngle, the face is considered a feature and one of its points will
be added to the coordinates used when refining the mesh.

doping
For EMW applications, doping is generally not required.
When the EnableEMW option is specified in the IOControls section of the command file,
doping is switched off to avoid unnecessary doping operations that may take too much CPU
time. If doping is required, the doping option can be used in the Mesh subsection to trigger
doping. By default, for typical applications, doping is switched on.

grading (1.25)
Specifies the grading in each direction. The default is 1.25 in each direction. This can be
specified in the following way:
• In three dimensions: grading = {gradx grady gradz}
• In two dimensions: grading = {gradx grady}

NOTE If the grading parameter is specified in both the Mesh subsection and
the EMW subsection, the EMW subsection takes precedence.

grading off
This statement switches off the grading refinement. By default, grading is switched on.

maxCellSize
Specifies the maximum cell size allowed in a region. The default cell size in each direction
is 10% of the geometry model size in that direction.

minCellSize (1e-4)
Specifies the minimum cell size (given in μm ) allowed in a region.

46 Sentaurus™ Mesh User Guide

O-2018.06
 2: Command File
Tensor Section

maxBndCellSize
Specifies the maximum cell size (given in μm ) perpendicular to each material interface on
the boundary. For this parameter, a normal vector is computed for each material interface
on the boundary, and a direction of the maximum projection is found. Cells are clustered
next to the material interface in the direction of the maximum projection. The default cell
size in each direction is 10% of the geometry model length in that direction.
In addition, you can restrict maxBndCellSize to an interface by specifying an interface
option as follows:
maxBndCellSize interface material | region "string" "string" float
To address external boundaries, you can use the keyword "Exterior" as one of the
materials or regions of an interface. For example:
maxBndCellSize interface material "Silicon" "Exterior" float
maxBndCellSize interface region "substrate" "Exterior" float
By default, if neither material "Exterior" nor region "Exterior" is specified, exterior
interfaces are not affected.

minBndCellSize (1e-4)
Specifies the minimum cell size (given in μm ) perpendicular to a material interface on the
boundary. For this parameter, a normal vector is computed for each material interface on
the boundary, and a direction of the maximum projection is found. The cell size next to the
material interface in the direction of the maximum projection will be, at least, the value of
minBndCellSize.
In addition, you can restrict minBndCellSize to an interface in the same way as for
maxBndCellSize.

minNumberOfCells (0)
Specifies the minimum number of cells required in each region and in each direction. The
actual number of cells is not necessarily the same as the value of minNumberOfCells due
to other parameters such as maxCellSize (default 10% of the entire structure) and
minCellSize (default 1e-4 μm ). The refinement algorithm for minNumberOfCells is
based on adaptive bisection, so cell sizes are not necessarily equidistant. If you want the
cell sizes to be equidistant, use maxCellSize.

numPoints
Specifies the fixed number of points in all directions.

numPointsX
Specifies the fixed number of points in the x-direction.

Sentaurus™ Mesh User Guide 47

O-2018.06
2: Command File
Tensor Section

numPointsY
Specifies the fixed number of points in the y-direction.

numPointsZ
Specifies the fixed number of points in the z-direction.

scale (1)
Specifies a mesh scaling factor. This parameter can be used to convert the mesh into
different units and can be specified in the following way:
• In three dimensions: scale = {sx sy sz}
• In two dimensions: scale = {sx sy}

window
Restricts the effects of the refinement parameters. The syntax for defining a window is the
following way:
• In three dimensions: window "windowname" xmin xmax ymin ymax zmin zmax
• In two dimensions: window "windowname" xmin xmax ymin ymax

xCuts, yCuts, zCuts

The values represent cuts in the tensor mesh where the refinement starts. These cuts are
kept as part of the final tensor mesh. The cuts in each direction are specified as a list of
double-precision values enclosed in parentheses. By default, no cuts are introduced into the
tensor mesh.

You can specify these refinement parameters (except for numPoints, numPointsX,
numPointsY, and numPointsZ) for a region, material, direction, or window, in any of the
following ways:

For all regions, in all directions parameter = floatOrint

In a region, in all directions parameter region "regionname" floatOrint

In a material, in all directions parameter material "materialname" floatOrint

In a window, in all directions parameter window "windowname" floatOrint

For all regions, in a direction parameter direction "x | y | z" floatOrint

In a region, in a direction parameter region direction "regionname" "x | y | z"

floatOrint

48 Sentaurus™ Mesh User Guide

O-2018.06
 2: Command File
Tensor Section

In a material, in a direction parameter material direction "materialname" "x | y | z"

floatOrint

In a window, in a direction parameter window direction "windowname" "x | y | z"

floatOrint

Sometimes, the same parameter is assigned a value multiple times, in which case, the last
assignment is taken into consideration.

Before specifying a parameter to be applied to a window, that window must be defined inside
a Mesh subsection of the command file. If a parameter specified through the window option
overlaps parameters specified with other options, the smallest of these parameters is considered
while meshing.

EMW Subsection for Computing Cell Size Automatically

When generating tensor meshes, the maximum cell sizes are computed automatically when the
EnableEMW option is specified in the IOControls section of the command file. This
application applies to Sentaurus Device Electromagnetic Wave Solver (EMW). The size
computed is a function of wavelength, nodes per wavelength, and the magnitude of a complex
refractive index (CRI).

The required parameters are specified in the EMW subsection of the Tensor section of the
command file. The syntax of the EMW subsection is:
Tensor {
EMW {
parameter filename = "string"
CRIMIPATH = "string"
CRIMODEL = "string"
CRI WavelengthDep Real Imag
grading = {float float float}
grading off
NoEMWSolverConstraintsCheck
wavelength = float
wavefrequency = float
CRI region "regionName" WavelengthDep Real Imag
CRI material "materialName" WavelengthDep Real Imag
CRI region "regionName" CRIMODEL "string"
CRI material "materialName" CRIMODEL "string"
npw | nodeperwavelength {
material "materialName" value
material direction "materialName" "x | y | z" float
region "regionName" float
region direction "regionName" "x | y | z" float
}

Sentaurus™ Mesh User Guide 49

O-2018.06
2: Command File
Tensor Section

npw | nodeperwavelength = integer

npwx | nodeperwavelengthX = integer
npwy | nodeperwavelengthY = integer
npwz | nodeperwavelengthZ = integer
}
}

where (default values are given in parentheses if applicable):

parameter filename = "string"

This statement sets the parameter filename that contains the CRI table of materials that
are present in the input structure.

CRIMIPATH
(Optional) Specifies the location of the CRI model.

CRIMODEL
(Optional) Specifies the name of the CRI model.

CRI WavelengthDep Real Imag

This statement sets the wavelength dependency on the real part, or the imaginary part, or
both parts of the CRI values. The specification of real and imaginary statements is optional.
Some examples are:
• Set the wavelength dependency only on the real part of the CRI:
CRI WavelengthDep Real
• Set the wavelength dependency only on the imaginary part of the CRI:
CRI WavelengthDep Imag
• Set the wavelength dependency on both the real and imaginary parts of the CRI:
CRI WavelengthDep

grading (1.25)
Specifies the grading in each direction. The default is 1.25 in each direction. This can be
specified in the following way:
• In three dimensions: grading = {gradx grady gradz}
• In two dimensions: grading = {gradx grady}

NOTE If the grading parameter is specified in both the Mesh subsection and
the EMW subsection, the EMW subsection takes precedence.

50 Sentaurus™ Mesh User Guide

O-2018.06
 2: Command File
Tensor Section

grading off
This statement switches off the grading refinement. By default, grading is switched on.

NoEMWSolverConstraintsCheck
Sentaurus Device Electromagnetic Wave Solver (EMW) has two limitations in handling
tensor meshes:
• It cannot handle tensor meshes with holes inside the structure.
• There must be at least one cell in each direction in a region.
By default, if these conditions are not met after generating the mesh, Sentaurus Mesh exits
with an error message.
During preprocessing, the tensor-product mesh generator attempts to determine whether
there will be at least one cell along a given direction inside a region before the tensor mesh
is constructed. This is not always possible, especially if regions have complicated
geometries.
This check is performed to save time before generating a tensor mesh, especially for large
structures.
If the tensor-product mesh generator detects a problem during preprocessing, it will
continue with mesh generation while issuing a warning message such as:
Warning: EMW applications require a minimum of 1 cell in each direction.
Region gate of material Oxide might not satisfy this condition in X direction.
Make sure that minCellSizeX for this region is at most 0.01
(or slightly smaller) to account for round-off errors.
You can do that by explicitly setting minCellSizeX, or through _numPointsX.
Using the NoEMWSolverConstraintsCheck option disables all of these checks.

wavelength (0.555)
Specifies the wavelength in micrometers.

wavefrequency
Specifies the value of the wavelength frequency. The wavelength is computed using this
value and the speed of light.

CRI region "regionName" WavelengthDep Real Imag

This statement sets the wavelength dependency for a specified region. Some examples are:
• Set the wavelength dependency for this region only on the real part of the CRI:
CRI region "Silicon_0" WavelengthDep Real

Sentaurus™ Mesh User Guide 51

O-2018.06
2: Command File
Tensor Section

• Set the wavelength dependency for this region only on the imaginary part of the CRI:
CRI region "Silicon_0" WavelengthDep Imag
• Set the wavelength dependency for this region on both the real and imaginary parts of
the CRI:
CRI region "Silicon_0" WavelengthDep

CRI material "materialName" WavelengthDep Real Imag

This statement sets the wavelength dependency for this material.

CRI region "regionName" CRIMODEL "string"

This statement sets a CRI model for a specified region.

CRI material "materialName" CRIMODEL "string"

This statement sets a CRI model for a specified material.

npw | nodeperwavelength (10)

This subsection defines the nodes per wavelength according to region or material, and in a
direction. If the direction is not used, the value is used in all directions. For example, the
following statement defines nodes per wavelength in silicon material in the x-direction:
npw { material direction "Silicon" "x" 20 }
The default value of the nodes per wavelength is 10 in all directions for each material.

For a given material, the cell size is computed using the formulas:
wavelength
λ mat = --------------------------- (1)
R mod

λ mat
cellsize mat = ---------- (2)
npw

The parameter R mod Eq. 1 is computed depending on the settings of the

ComplexRefractiveIndex model:

• If WavelengthDep Real is used, R mod = n .

• If WavelengthDep Imag is used, R mod = k .
2 2
• If WavelengthDep Real Imag is used, R mod = n +k .
Here, n is the real part and k is the imaginary part of the CRI.

52 Sentaurus™ Mesh User Guide

O-2018.06
 2: Command File
Tensor Section

npwx | nodeperwavelengthX
Sets the nodes per wavelength similar to npw but only in the x-direction for all materials.

npwy | nodeperwavelengthY
Sets the nodes per wavelength similar to npw but only in the y-direction for all materials.

npwz | nodeperwavelengthZ
Sets the nodes per wavelength similar to npw but only in the z-direction for all materials.

Box Subsection for Plotting

New regions can be added to the tensor mesh that can be used for plotting purposes in EMW
applications. The new regions are specified by the Box subsection of the Tensor section of the
command file.

Any number of Box subsections can be specified in the Tensor section of the command file.
The Box subsections are added outside the Mesh subsection.

The syntax of the Box subsection is:

Tensor {
Box {
boundingBox
boundingBox region = "string"
endPoint = {float float float}
exact = "yes" | "no"
material = "string"
name = "string"
startPoint = {float float float}
tolerance = float
}
}

where (default values are given in parentheses if applicable):

boundingBox
This option can be used instead of specifying startPoint and endPoint. It
automatically sets the minimum and maximum of the structure bounding box as the
startPoint and endPoint, respectively.

Sentaurus™ Mesh User Guide 53

O-2018.06
2: Command File
Tools Section

boundingBox region = "string"

This statement can be used instead of specifying startPoint and endPoint. It
automatically sets the minimum and maximum of the region bounding box as the
startPoint and endPoint, respectively.

endPoint
Specifies the highest point of the bounding box used for the plot (xmax ymax zmax).

exact ("no")
If exact="yes", the resultant mesh should contain nodes whose coordinates match
startPoint and endPoint. If exact="no", the nodes that are closest to startPoint
and endPoint are written in the tensor mesh.

material ("none")
If not specified, the name of the material defaults to "none".

name
Name of this region.

startPoint
Specifies the lowest point of the bounding box used for the plot (xmin ymin zmin).

tolerance
The tolerance is used only if exact="yes". The tolerance value indicates that the box
should be aligned to any existing boundary or cell interface within this tolerance distance.
This avoids unnecessary small cells locally. A value of model length in each direction
multiplied by 1e-4 μm is used as the default.
In two dimensions, tolerance= {tx, ty}.
In three dimensions, tolerance= {tx, ty, tz}.

Tools Section
The Tools section is used to execute geometric operations on either a boundary file or a mesh
file. The input mesh can be either a tetrahedral mesh or a hybrid (mixed-element) mesh.

If a hybrid mesh is used, it must be converted to a tetrahedral mesh before applying the tool
(see Converting a Tetrahedral Mesh to a Hybrid Mesh on page 63). The mesh is converted back
to a hybrid mesh after all operations have been executed. If an operation such as a simple

54 Sentaurus™ Mesh User Guide

O-2018.06
 2: Command File
Tools Section

transformation is applied, the resulting mesh might differ slightly from the original mesh,
despite no topological changes.

The operations are executed according to their order in the Tools section. The output of one
operation becomes the input for the next operation.

The syntax of the Tools section is:

Tools {
parameters
}

NOTE To use the Tools section, you must specify the EnableTools option
in the IOControls section of the command file (see IOControls
Section on page 6).

Appending the Input Structure

This section appends the input structure periodically at the specified position:
Tools {
Append {
axis = xmin | ymin | zmin | xmax | ymax | zmax
map "stringA" = "stringB"
}
}

NOTE It works only for 2D and 3D boundaries.

Creating Profiles
This section creates profiles in the input mesh with the description given in the command file:
Tools {
CreateProfiles {
SrcMesh = "string"
CmdFile = "string"
}
}

The mesh is not modified in this process as the refinement specifications in the command file
are ignored. During profile creation, the existing profiles in the input mesh, which are again
specified in the command file, will only be recreated. The rest of the profiles are untouched.

Sentaurus™ Mesh User Guide 55

O-2018.06
2: Command File
Tools Section

Setting a Transformation
This section sets a transformation matrix to a mesh or a boundary:
Tools {
Set Transformation {
translation = (float float [float])
scale = float | scale = (float float float)

rotation {
axis = (float float float)
angle = float
} |
rotation {
matrix (float float float float float float float float float)
}
}

Apply Transformation
}

NOTE The Set Transformation operation applies to both the mesh and
boundary.

A translation vector is used to set a translation. The rotation can be set by either an axis vector
and an angle in degrees, or a matrix. A mesh or a boundary also can be scaled by specifying a
floating-point value or a vector.

The Apply Transformation statement applies the transformation that is set.

Removing Short Features

This section removes unwanted short features in a boundary:
Tools {
Decimate {
accuracy = float
shortedge = float
}
}

The accuracy parameter indicates the deviation of a structure from its original location. The
default is 1e-8.

56 Sentaurus™ Mesh User Guide

O-2018.06
 2: Command File
Tools Section

The shortedge parameter removes short edges. All edges that are shorter than this parameter
are eliminated. This parameter does not have a default value and is activated only when it has
a nonzero value.

Rediscretizing the Boundary Using the DelPSC Algorithm

This section rediscretizes the boundary by surface remeshing using the DelPSC algorithm that
creates good-quality triangles on non-flat surfaces of the boundary:
Tools {
DelaunaySurfaceRemeshing {
DelPSCAccuracy = float
DelPSCRidgeAngle = float
DelPSCRidgeSampling = float
}
}

The following parameters control the quality of the discretization:

■ DelPSCAccuracy controls the deviation between the new curved surfaces and the original
curved surfaces. The new curved surface can deviate from the original curved surface by,
at most, the value of DelPSCAccuracy. New vertices lie exactly on the original surface,
but new triangles cannot lie exactly on the original surface unless the original surface is flat.
In general, the smaller the value of DelPSCAccuracy is, the smoother the new surface
becomes, and the more accurate the new surface represents the original surface.
■ DelPSCRidgeAngle is an angle computed at each edge. It is a dihedral angle between two
shared faces. This parameter identifies the geometric features. Default is 95° .
■ DelPSCRidgeSampling discretizes the ridges and controls the size of small triangles.
Default is 0.01 μm .

NOTE The DelPSC algorithm uses multithreading if you specify

IOControls{numThreads=integer}.

Rediscretizing the Boundary Using the Dual-Contouring

Algorithm
This section rediscretizes the boundary by surface remeshing using the dual-contouring
algorithm. Dual contouring can be useful when the input boundary is low quality, that is, it
contains triangles with small planar or dihedral angles. Dual contouring preserves sharp edges
as much as possible and avoids generating small features that cannot be meshed. See [3][4][5]
for a description of the algorithm.

Sentaurus™ Mesh User Guide 57

O-2018.06
2: Command File
Tools Section

Tools {
DualContouring {
decimation = true | false
minAngle = float
minDihedralAngle = float
resolution = float
}
}

The following parameters control the quality of the discretization:

■ decimation applies a decimation algorithm to reduce the number of nodes on the
rediscretized boundary. Default: false.
■ minAngle is the minimum planar angle on the boundary after rediscretization. Default: 5°.
■ minDihedralAngle is the minimum dihedral angle on the boundary after
rediscretization. Default: 10°.
■ resolution defines the minimum spacing of an octree grid used for the boundary
rediscretization. In general, the smaller the value of resolution, the more accurately the
new surface represents the original surface.

NOTE The dual-contouring algorithm uses multithreading if you specify

IOControls{numThreads=integer}.

Interpolating a Source Mesh to a Destination Mesh

This section allows interpolation from the source mesh to the destination mesh:
Tools {
InterpolateMesh {
DstMesh = "string"
Conservative
Extrapolate = true | false
IgnoreMaterials
Species {"string" "string" ...}
SrcMesh = "string"
Tolerance = float
}
}

The SrcMesh parameter specifies the file name of the source mesh, and the DstMesh
parameter specifies the file name of the destination mesh. By default, if no species is specified,
all the fields in the source mesh are considered for interpolation.

If the Conservative option is specified, a second-order method described in [6] is used to

perform the interpolation. If any species is specified inside the Species subsection, only those

58 Sentaurus™ Mesh User Guide

O-2018.06
 2: Command File
Tools Section

species will be interpolated. If the species is already present in the destination mesh, the values
will be overwritten with new interpolated values. By default, the interpolation is performed
between two identical materials, and the IgnoreMaterials option can be used to override
this behavior.

When the boundary of the source and the destination meshes do not coincide exactly,
Extrapolate=true (default) will perform the extrapolation by assigning the field value for
the destination point from the closest point on the source mesh. For some applications,
extrapolation is not wanted, in which case, set Extrapolate=false and set the appropriate
Tolerance for searching for source elements that contain destination points.

Performing a 2D Slice of 3D Mesh or Boundary

This section performs a 2D slice of a 3D mesh or boundary:
Tools {
Slice {
location = (float float float)
normal = (float float float)
} |
Slice {
Direction = X | Y | Z
Endpoint = (float float)
Startpoint = (float float)
}
}

NOTE The Slice operation applies to both the mesh and boundary.

The 2D slice is defined by a plane normal and a location (see Slicing a 3D Mesh Using a Plane
and Its Location on page 105).

Alternatively, the slice can be obtained by restricting a plane by a segment. For this interface,
a direction parameter indicating the plane in which the segment lies, and a starting point and
an endpoint of a segment are needed. The starting point and endpoint are represented by only
two coordinates of the segment, and the third coordinate is computed from the input structure.
For example, if the direction is the z-plane, the coordinates of Startpoint and Endpoint
represent the x- and y-values of the segment. If the direction is the y-plane, the coordinates of
Startpoint and Endpoint represent the x- and z-values of the segment. Similarly, if the
direction is the x-plane, the coordinates of Startpoint and Endpoint represent the y- and
z-values of the segment.

Sentaurus™ Mesh User Guide 59

O-2018.06
2: Command File
Tools Section

Cutting a Mesh With a Plane

This section cuts a mesh with a plane:
Tools {
Cut {
normal = (float float float)
location = (float float float)
}
}

The mesh that is to the right of the plane is removed. The right side of the plane is defined as
the one to which the normal points.

NOTE This operation is limited to 3D meshes and 3D boundaries. For hybrid

meshes, cutting through elements will result in significant topological
changes around the cutplane.

Reflecting a Mesh
This section mirrors a mesh about a location and appends it to the original mesh:
Tools {
Reflection {
axis = xmin | ymin | zmin | xmax | ymax | zmax
map "stringA" = "stringB"
}
}

NOTE This operation is limited to meshes.

The map statement specifies the name that corresponds an input region to the mirrored region.
By default, if map is not specified, the new region names are given the _mirrored file
extension.

60 Sentaurus™ Mesh User Guide

O-2018.06
 2: Command File
Tools Section

Extruding a Mesh
This section extrudes a planar 2D mesh along a nonuniform 1D grid to generate a 3D mesh:
Tools {
Extrude {
spacingMethod = even | regular | smooth
zCuts = (floatlist)
}
}

The following parameters control extrusion of a mesh:

■ spacingMethod specifies the type of progression used when a sequence of cutplanes is
generated between adjacent pair of planes. The available options are:
• even distributes the cuts evenly, trying to approximate the spacing specified at the
beginning of the interval.
• regular distributes the cuts evenly, using the exact spacing specified at the beginning
of the interval, and leaves the last interval with an approximate size if there is no more
room to accommodate the requested size.
• smooth distributes the cuts to have a smooth grading of spacing between planes.

Default: even.
■ zCuts represents the cutplanes for extrusion. Each cutplane can be either a single floating-
point number that represents the distance of the cut parallel to the 2D planar mesh or a pair
of floating-point numbers. In the latter case, the first number indicates the distance of the
cut from the 2D mesh, and the second number indicates the expansion of the cuts into a
sequence of planes to be generated between adjacent pair of planes.

For example, zCuts = (0.2 0.5 0.8) generates three cutplanes at distances 0.2, 0.5, and
0.8 μm from the input 2D mesh.

The following example creates a series of cutplanes at distances 1.2, 2.5, and 2.9 μm from the
input 2D mesh:
zCuts = ( (1.2 0.3) (2.5 0.2) (2.9) )

However, in this case, between 1.2 and 2.5 μm , additional cutplanes are generated within
0.3 μm intervals. Between 2.5 and 2.9 μm , additional cutplanes are generated within 0.2 μm
intervals (depending on the value of the spacingMethod parameter).

Sentaurus™ Mesh User Guide 61

O-2018.06
2: Command File
Tools Section

Stretching a Mesh
This section stretches an existing mesh by adding a new column of elements at the specified
location in the specified direction:
Tools {
Stretch {
direction = X | Y | Z
length = float
location = (float float float)
}
}

The unit of length is the same as that of the input mesh. A negative length indicates a stretch in
the opposite direction. This operation is limited to meshes.

Placing Individual Dopant of Species

This section allows you to place an individual dopant of a species at a specified location:
Tools {
Dopants {
Species "string" {
DopantLocation = (float float float)
...
DopantLocation = (float float float)
Replace
}
}
}

If the specified location matches the mesh point, the discrete dopant will be assigned to that
particular point or it will snap to the nearest mesh point. By default, the specified dopants for
each species are added to the input continuous doping. If you specify the Replace option, the
input continuous doping is reinitialized to zero for a user-specified species only in those
regions that contain the specified single dopants. The other species of this region are
untouched.

For example, if you specify single dopants of BoronActiveConcentration with the

Replace option, only the input BoronActiveConcentration of the region that contains
these single dopants is reinitialized to zero, and the contribution from single dopants is
considered in that region. The other species of this region are untouched.

62 Sentaurus™ Mesh User Guide

O-2018.06
 2: Command File
Tools Section

Extracting a Boundary From a Mesh

You can specify the Mesh2bnd option to extract a boundary from a mesh, or you can set the
geometric accuracy and short edge to clean up unwanted small features of the input geometry:
Tools {
Mesh2bnd |

Mesh2bnd {
accuracy = float
shortedge = float
}
}

The default value of accuracy is 1e-8 μm . The geometric accuracy cleans up the coplanar
mesh points. By default, this operation does not remove the short edges.

Converting a Tetrahedral Mesh to a Hybrid Mesh

Hybrid meshes (also referred to as mixed-element meshes) contain hexahedra, prisms,
pyramids, and tetrahedra. They are used in some tools such as Sentaurus Device because,
compared to tetrahedral meshes, hybrid meshes have fewer elements, thereby allowing the
tools to perform simulations on larger structures. Hybrid meshes are also used in Sentaurus
Interconnect to increase the accuracy of the simulation.

The Mesh2Hybrid option converts an input mesh containing only tetrahedral elements to a
hybrid mesh:
Tools {
Mesh2Hybrid
}

NOTE Sentaurus Mesh cannot use hybrid meshes directly as input for any
operation specified in the Tools section.

When the Mesh2Hybrid option is processed, the output is written out

directly, thereby ignoring the operations that follow in the Tools
section.

Sentaurus™ Mesh User Guide 63

O-2018.06
2: Command File
Tools Section

Specifying Algorithm for Smoothing Noise

This section uses a multimaterial level-set (MLS) algorithm to smooth any noise that may be
present in the boundary file:
Tools {
MultiLevelSetBrepFilter {
CellSize = float
numThreads = integer
}
}

The output of this section may contain a large number of poor-quality triangles on non-flat
surfaces. You should use the DelaunaySurfaceRemeshing section afterwards to eliminate
poor-quality triangles (see Rediscretizing the Boundary Using the DelPSC Algorithm on
page 57).

The CellSize parameter specifies the level-set cell size, which should be, at most, one-third
the thickness of the thinnest region. Otherwise, the thin region may be considered noise and it
disappears. The coarser the cell size, the more features may be smoothened. Default is
0.001 μm .

The amount of geometry smoothing performed by the MLS algorithm depends on both
curvatures in the input and the level-set cell size. A noisy surface has a high curvature, so it will
be smoothed to a large extent to remove noise. On the other hand, a planar surface has zero
curvature, so it will be well preserved. Unfortunately, a sharp corner has a theoretically infinite
curvature, so it will become a rounded corner. The specified level-set cell size is the threshold
to distinguish between the noise to be removed and the features to be preserved.

The numThreads parameter specifies the number of threads to use. Default is 1.

Creating Structures With Randomized Doping Profiles

This section creates structures with randomized doping profiles based on an original structure
obtained from process simulation or created analytically. The section works by atomizing the
original continuous doping distribution to create discrete dopants and then reassigning the
doping associated with these discrete dopants back to the surrounding mesh nodes. Different
atomizations or randomized structures can be obtained from one original structure.

NOTE This section does not change the mesh but reassigns the randomized
doping.

64 Sentaurus™ Mesh User Guide

O-2018.06
 2: Command File
Tools Section

The syntax is:

Tools {
RandomizeDoping {
ContinuousContactDoping
DopingAssignment = "Sano" | "CIC" | "NGP"
FileIndex = integer
NumberOfRandomizedProfiles = integer
SaveDiscreteDopants

Material "Material name 1" {

Species "Dataset name 1" {
Ignore | Randomize
ScreeningFactor = value
AutoScreeningFactor
}
Species "Dataset name 2" {
Ignore | Randomize
ScreeningFactor = value
AutoScreeningFactor
}
...
}

Material "Material name 2" {

Species "Dataset name 1" {
Ignore | Randomize
ScreeningFactor = value
AutoScreeningFactor
}
Species "Dataset name 2" {
Ignore | Randomize
ScreeningFactor = value
AutoScreeningFactor
}
...
}

Region "Region name 1" {

Species "Dataset name 1" {
Ignore | Randomize
ScreeningFactor = value
AutoScreeningFactor
}
Species "Dataset name 2" {
Ignore | Randomize
ScreeningFactor = value
AutoScreeningFactor
}

Sentaurus™ Mesh User Guide 65

O-2018.06
2: Command File
Tools Section

...
}

Cuboid [ (value value value) (value value value) ] {

Species "Dataset name 1" {
ScreeningFactor = value
ScreeningScalingFactor = value
AutoScreeningFactor
}
Species "Dataset name 2" {
ScreeningFactor = value
ScreeningScalingFactor = value
AutoScreeningFactor
}
...
}
...
}
}

where:

ContinuousContactDoping
Specifying this option discards the randomized doping assigned to the contact nodes and,
instead, it uses the contact doping obtained from the original structure.

DopingAssignment = "Sano" | "CIC" | "NGP"

Specifies the method used to assign doping from the discrete dopants created during
atomization to the mesh nodes:
• The cloud-in-cell ("CIC") method distributes the doping of a particle to the vertex
nodes of the element in which the particle is located.
• The nearest grid point ("NGP") method assigns the doping of a particle to the nearest
mesh node.
• The "Sano" method uses a doping function described in Appendix B on page 129 to
distribute the doping of a particle to surrounding nodes.

FileIndex (0)
Specifies the name of output files and also is used as a random seed during randomization.
The names of output files are created automatically using the following convention:
<root>_<DopingAssignment><FileIndex + Randomized_Profile_Number>_msh.tdr
where <root> is a base name of the input TDR file.

66 Sentaurus™ Mesh User Guide

O-2018.06
 2: Command File
Tools Section

For example, if DopingAssignment = "Sano", FileIndex = 1000, and

NumberOfRandomizedProfiles = 3, and the command line is:
snmesh nmos
then the following output files are created:
nmos_sano1000_msh.tdr
nmos_sano1001_msh.tdr
nmos_sano1002_msh.tdr

NumberOfRandomizedProfiles (1)
Specifies the number of randomized profiles to be generated.

SaveDiscreteDopants
This option saves active discrete dopants along with the randomized dopant profiles. The
active discrete dopants can be visualized as well as the mesh. This file also can be specified
as a ParticleFile in the particle profile section of the Definitions section (see
Defining Particle Profiles on page 17).

Ignore | Randomize
Two choices are provided for each specified species. By default, all specified species are
randomized. With the Ignore option, a species is not randomized and is not copied to the
output file. All species that are not specified in the command file for a specified material
are copied to the output file.

In the RandomizeDoping section, only materials that are specified in the command file will
have their doping randomized. Materials found in the input structure that are not specified in
the command file will have their original continuous doping written to the output file. If a
species is not specified for a material, it is simply copied to the output without randomizing.

For example, if the input structure contains the material "Silicon" with the species
"ArsenicActiveConcentration" and "BoronActiveConcentration", and the
command file only specifies "BoronActiveConcentration", the randomized species
"BoronActiveConcentration" and original "ArsenicActiveConcentration" will
contribute to the silicon doping in the final output structure.

The ScreeningFactor parameter and the AutoScreeningFactor option are used only
with DopingAssignment = "Sano". If "Sano" is selected, ScreeningFactor must be
specified. The AutoScreeningFactor calculation is invoked only if specified.

In addition to materials, randomization can be restricted solely to a region or a cuboid.

Sentaurus™ Mesh User Guide 67

O-2018.06
2: Command File
QualityReport Section

NOTE When the RandomizeDoping section is processed, the output is

written out directly, thereby ignoring the operations that follow in the
Tools section.

Adding or Removing Interfaces From a Mesh

Some TCAD Sentaurus tools produce meshes that contain an explicit description of the
interface between adjacent regions. This description is not understood by some tools and is not
produced by other tools, so it is sometimes necessary to add or remove it from a mesh file.

You can specify either the addInterfaceRegions or the removeInterfaceRegions

option to add interfaces or to remove interfaces, respectively.

The syntax is:

Tools {
addInterfaceRegions | removeInterfaceRegions
}

QualityReport Section
The QualityReport section is optional and is used to specify mesh quality limits for mesh
generation. Sentaurus Mesh produces a report regardless of whether the limits are satisfied or
not. This section of the command file can help to ensure that the mesh is suitable for device
simulation.

NOTE The QualityReport section applies only to 3D axis-aligned meshes

and 3D offsetting meshes. The specified limits are used only to report
on the mesh quality and do not affect how meshes are generated.

If any limits are not satisfied, Sentaurus Mesh saves additional field variables in the output
_msh.tdr file:
■ AngleElements: The angle of an element as defined by the box method.
■ DelaunayInsphere3D: The number of elements that are non-Delaunay elements.
■ ElementsPerVertex: The number of elements that share a vertex.
■ ElementVolume: The volume of an element.
■ ShortestEdge: The length of the shortest edge of an element.

68 Sentaurus™ Mesh User Guide

O-2018.06
 2: Command File
QualityReport Section

The syntax of this section is:

QualityReport {
Global
Material = {stringList}
Region = {stringList}
{
limitMaxConnectivity = integer
limitMaxNonDelaunay = float
limitMinAngle = float
limitMinEdgeLength = float
limitMinVolume = float
}
}

where (default values are given in parentheses if applicable):

Global
If specified, the limits are evaluated on the entire mesh.

Material
If specified, the limits are evaluated on the list of materials only.

Region
If specified, the limits are evaluated on the list of regions only.

limitMaxConnectivity (0)
Specifies the maximum number of elements that can share any vertex. If this limit is
exceeded, Sentaurus Mesh saves the ElementsPerVertex field variable in the output
mesh file. The default value of 0 means this parameter has no effect.

limitMaxNonDelaunay (100.0)
Specifies the maximum percentage of all elements that can be non-Delaunay elements. If
this limit is exceeded, Sentaurus Mesh saves the DelaunayInsphere3D field variable in
the output mesh file.

limitMinAngle (0.0)
Specifies the minimum angle (given in degrees), defined using the box method, of any
element. If this limit is exceeded, Sentaurus Mesh saves the AngleElements field
variable in the output mesh file. See Utilities User Guide, AngleElements on page 32.

Sentaurus™ Mesh User Guide 69

O-2018.06
2: Command File
QualityReport Section

limitMinEdgeLength (0.0)
Specifies the minimum edge length (given in μm ) of any element. If this limit is exceeded,
Sentaurus Mesh saves the ShortestEdge field variable in the output mesh file.

limitMinVolume (0.0)
3
Specifies the minimum volume (given in μm ) of any element. If this limit is exceeded,
Sentaurus Mesh saves the ElementVolume field variable in the output mesh file.

Examples

Generate a report on the mesh quality of the entire mesh using the default limits:
QualityReport {
Global
}

Generate a report on the mesh quality of the entire mesh using the default limits, followed by
a report on the mesh quality of the materials Silicon and Oxide using the default limits:
QualityReport {
Global
Material = {"Silicon" "Oxide"}
}

Generate a report on the mesh quality of the entire mesh using the default limits, followed by
a report on the mesh quality of the regions Substrate and Oxide_1 using the default limits:
QualityReport {
Global
Region = {"Substrate" "Oxide_1"}
}

Generate a report on the mesh quality of the entire mesh with two specific limits:
QualityReport {
Global {
limitMaxNonDelaunay = 0.1
limitMinAngle = 1e-2
}
}

Generate a report on the mesh quality of the entire mesh with one set of limits, followed by a
report on the mesh quality of the material Silicon with a different set of limits:
QualityReport {
Global {
limitMaxNonDelaunay = 0.1
limitMinAngle = 1e-2

70 Sentaurus™ Mesh User Guide

O-2018.06
 2: Command File
References

}
Material = {"Silicon"} {
limitMinVolume = 1e-18
limitMinEdgeLength = 1e-5
}
}

References
[1] S.-W. Cheng, T. K. Dey, and J. A. Levine, “A Practical Delaunay Meshing Algorithm
for a Large Class of Domains,” in Proceedings of the 16th International Meshing
Roundtable, Seattle, WA, USA, pp. 477–494, October 2007.
[2] T. K. Dey and J. A. Levine, “Delaunay Meshing of Piecewise Smooth Complexes
without Expensive Predicates,” Algorithms, vol. 2, no. 4, pp. 1327–1349, 2009.
[3] T. Ju et al., “Dual Contouring of Hermite Data,” ACM Transactions on Graphics (TOG),
vol. 21, no. 3, pp. 339–346, 2002.
[4] S. Schaefer and J. Warren, Dual Contouring: “The Secret Sauce”, Technical Report,
Department of Computer Science, Rice University, Houston, TX, USA, March 2003.
[5] Y. Zhang and J. Qian, “Dual Contouring for domains with topology ambiguity,”
Computer Methods in Applied Mechanics and Engineering, vol. 217–220, pp. 34–45,
April 2012.
[6] F. Alauzet and M. Mehrenberger, “P1-conservative solution interpolation on
unstructured triangular meshes,” International Journal for Numerical Methods in
Engineering, vol. 84, no. 13, pp. 1552–1588, 2010.

Sentaurus™ Mesh User Guide 71

O-2018.06
2: Command File
References

72 Sentaurus™ Mesh User Guide

O-2018.06
CHAPTER 3 Doping and Refinement Examples

This chapter illustrates how to use the Definitions and Placements

sections of the command file, as well as how to use the Offsetting
section to generate layered meshes in Sentaurus Mesh.

Command File for a Simple Diode

This section describes the command file diode.cmd that is used as an example for the rest of
this chapter. The command file contains two types of information: dimension-independent data
and dimension-dependent data. The dimension-independent part of the command file
diode.cmd, for this example, is:
Title "minimal example: simple diode"
Definitions {
# Profiles
Constant "n-type region" {
Species = "PhosphorusActiveConcentration" Value = 1e+18
}
Constant "p-type region" {
Species = "BoronActiveConcentration" Value = 1e+17
}
}

The optional keyword Title is used for a short description of the device and mesh. The
Definitions section specifies the dimension-independent part of the command file and can
be used for all dimensions without modifications.

Two constant profiles for doping are described using the keyword Constant followed by the
profile name in double quotation marks. The keyword Species is used to declare the doping
species used in the region. The constant concentration is specified by the number following the
keyword Value. The sign is intrinsic to the species.

Now, the doping profiles must be placed in the device. The placement of these profiles depends
on the device geometry. Since, for this example, ‘solid’ regions are to be filled with constant
doping, these instructions are added to the command file:
Placements {
# Profiles
Constant "n-type region instance" {
Reference = "n-type region"
EvaluateWindow {

Sentaurus™ Mesh User Guide 73

O-2018.06
3: Doping and Refinement Examples
Refinement and Evaluation Windows

Element = cuboid [(1 0 0), (2 3 2)] # for 3D

}
}
Constant "p-type region instance" {
Reference = "p-type region"
EvaluateWindow {
Element = cuboid [(0 0 0), (1 3 2)] # for 3D
}
}
}

The keyword Placements starts the dimension-dependent section where the instances of the
definitions given in the Definitions section are defined. The Reference parameter
specifies a profile defined in the Definitions section. The EvaluateWindow defines the
valid domain for the profiles. In this example, the valid domains are lines in 1D, rectangles in
2D, and cuboids in 3D. If EvaluateWindow is not defined in the file, the profile is valid in the
entire domain of the device.

For the 3D case, the valid domain of the p-type region is the lower half of the device, given by
the cuboid [(0 0 0), (1 2 3)]. In 2D, this domain is given by the rectangle [(0 0), (1 2)] and in
1D, by the line [(0), (1)]. However, the doping profile defined for 3D can be used for the lower
dimensions and, for the rest of this chapter, only the command file for the 3D case will be used.

In the example, there is an abrupt decay function between the two constant profiles. The doping
associated with points outside the EvaluateWindow is zero. This situation can be modified if
the parameter DecayLength is used. By setting DecayLength in EvaluateWindow, an error
function can be used as a decay profile.

Refinement and Evaluation Windows

The refinement conditions specified inside a Refinement statement can be restricted using
refinement windows. The windows can be simple rectangles, polygons, polyhedra, regions, or
materials.

Using Refinement Polygons

Figure 1 on page 76 illustrates the use of polygonal domains for specifying a polygonal
RefineWindow and for using a polygonal domain as an EvaluateWindow. The domain is a
simple rectangular boundary and the command file is:
Title "Refinement Polygon"
Definitions {
Refinement "global" {

74 Sentaurus™ Mesh User Guide

O-2018.06
 3: Doping and Refinement Examples
Refinement and Evaluation Windows

MaxElementSize = (4, 4)
MinElementSize = (.04 .04)
RefineFunction = MaxTransDiff(Variable="DopingConcentration", Value=0.5)
}
Refinement "refpol" {
MaxElementSize = (0.3 0.1)
}
Constant "bor" {
Species = "BoronConcentration" Value=1e+17
}
}
Placements {
Refinement "global" {
Reference = "global"
RefineWindow = rectangle [( -2 -2 ), ( 14 14 )]
}
Refinement "refpol" {
Reference = "refpol"
RefineWindow = polygon [( 1 2 ) ( 0.75 2 ) ( 1 2.5 ) ( 1.25 3 )
( 1.5 3.5 ) ( 1.75 4 ) ( 2 4.25 ) ( 2.25 4.5 ) ( 2.5 4.75 ) ( 2.75 5 )
( 2.75 5.5 ) ( 3 5.75 ) ( 3.5 5.5 ) ( 4 5.75 ) ( 4.5 5.5 ) ( 5 5.5 )
( 5.5 5.75 ) ( 5.5 6 ) ( 6 6.25 ) ( 6.5 6 ) ( 7 6 ) ( 7.5 5.25 )
( 8 5.5 ) ( 8 5 ) ( 7.5 4.5 ) ( 8 4.25 ) ( 8.5 4 ) ( 9 3.75 ) ( 9.5 4 )
( 9.5 3.5 ) ( 9.5 3 ) ( 9 3 ) ( 8.5 2.75 ) ( 8.75 2.5 ) ( 8.5 2.25 )
( 8 2.25 ) ( 7.5 2.25 ) ( 7.5 2.5 ) ( 7 2.5 ) ( 7 2 ) ( 6.75 1.5 )
( 6.75 1 ) ( 6.25 1 ) ( 6 1.5 ) ( 5.5 2 ) ( 5.5 2.5 ) ( 5 2 )
( 4.75 1.5 ) ( 4.5 1 ) ( 4 1.25 ) ( 3.5 1.25 ) ( 3 1 ) ( 2.5 1.5 )
( 2.5 2 ) ( 2.5 2.5 ) ( 2 2.5 ) ( 1.5 2.5 ) ( 1.5 2 ) ( 1 2 )]
}
Constant "bor" {
Reference = "bor"
EvaluateWindow {
Element = polygon [( 1 2 ) ( 0.75 2 ) ( 1 2.5 ) ( 1.25 3 ) ( 1.5 3.5 )
( 1.75 4 ) ( 2 4.25 ) ( 2.25 4.5 ) ( 2.5 4.75 ) ( 2.75 5 )
( 2.75 5.5 ) ( 3 5.75 ) ( 3.5 5.5 ) ( 4 5.75 ) ( 4.5 5.5 ) ( 5 5.5 )
( 5.5 5.75 ) ( 5.5 6 ) ( 6 6.25 ) ( 6.5 6 ) ( 7 6 ) ( 7.5 5.25 )
( 8 5.5 ) ( 8 5 ) ( 7.5 4.5 ) ( 8 4.25 ) ( 8.5 4 ) ( 9 3.75 )
( 9.5 4 ) ( 9.5 3.5 ) ( 9.5 3 ) ( 9 3 ) ( 8.5 2.75 ) ( 8.75 2.5 )
( 8.5 2.25 ) ( 8 2.25 ) ( 7.5 2.25 ) ( 7.5 2.5 ) ( 7 2.5 ) ( 7 2 )
( 6.75 1.5 ) ( 6.75 1 ) ( 6.25 1 ) ( 6 1.5 ) ( 5.5 2 ) ( 5.5 2.5 )
( 5 2 ) ( 4.75 1.5 ) ( 4.5 1 ) ( 4 1.25 ) ( 3.5 1.25 ) ( 3 1 )
( 2.5 1.5 ) ( 2.5 2 ) ( 2.5 2.5 ) ( 2 2.5 ) ( 1.5 2.5 ) ( 1.5 2 )
( 1 2 )]
}
}
}

Sentaurus™ Mesh User Guide 75

O-2018.06
3: Doping and Refinement Examples
Refinement and Evaluation Windows

Figure 1 Polygonal refinement

Using Composite Elements

Geometric elements can be combined to form more complex elements. This can be used to
define curved reference elements for analytic profiles, which otherwise cannot be correctly
defined using the standard elements. An example of the use of a composite element is:
AnalyticalProfile "MyProfile" {
Reference = "MyProfileReference"
ReferenceElement {
Element = {
line [( -1 2 ) , ( 4 2 )]
line [( 4 2 ) , ( 6 4 )]
line [( 6 4 ) , ( 7 4 )]
}
}
}

NOTE For the composite element to be effective, all components must be

adjacent to each other, without leaving gaps between them.

Composite elements are available only in Sentaurus Mesh.

76 Sentaurus™ Mesh User Guide

O-2018.06
 3: Doping and Refinement Examples
Refinement and Evaluation Windows

-1

-0.5
Y [µm]

0.5

-1 -0.5 0 0.5 1
X [µm]
Figure 2 Profile built by combining several reference elements

Regionwise and Materialwise Refinement

Figure 3 on page 78 illustrates the effect of using regionwise and materialwise refinement. The
following command file segment shows the relevant part of the command file:
Placements {
Refinement "A" {
Reference = "A"
RefineWindow = region ["Ox_Region"]
}
Refinement "B" {
Reference = "B"
RefineWindow = material ["Oxide"]
}
}

Sentaurus™ Mesh User Guide 77

O-2018.06
3: Doping and Refinement Examples
Using Analytic Functions for Doping Specification

Figure 3 (Left) Regionwise and (right) materialwise refinement

Using Analytic Functions for Doping Specification

This example illustrates the use of general analytic functions for defining doping profiles. To
use the primary and lateral directions as x and y, the keyword Eval must be specified (instead
of General), that is, by using global spatial coordinates. Figure 4 shows the generated meshes.
AnalyticalProfile "NoName_0" {
Species = "BoronActiveConcentration"
Function = Eval(init="a=10",function = "a*sin(x)*cos(y)",value = 0)
}
ReferenceElement {
Element = line [( 0 0 ), ( 10 10 )]
# Element = line [( 0 5 ), ( 10 5 )]
# Element = line [( 5 0 ), ( 5 10 )]
}

Element=line[(0 0),(10 10)] Element=line[(0 5),(10 5)] Element=line[(5 0),(5 10)]

Figure 4 Use of analytic refinement functions for doping

78 Sentaurus™ Mesh User Guide

O-2018.06
 3: Doping and Refinement Examples
Creating 3D Profiles From 2D Cross Sections

Creating 3D Profiles From 2D Cross Sections

A 2D profile can be extended into three dimensions by using an EvaluateWindow containing
a sweepElement, which is an advanced type of element that allows a 2D geometric element
to be either swept along a path or swept about a reference axis.

A sweepElement is composed of a base element, which can be either a polygon or a rectangle,

and a path, which can be represented in several ways. When a 3D profile is evaluated, the
sweepElement takes the 3D coordinate and follows the path in reverse, calculating a local 2D
coordinate on the base element. This 2D coordinate is then used to evaluate the 2D profile and
to provide the values in 3D.

NOTE A sweepElement can be used only in EvaluateWindow statements.

Using it in a ReferenceElement or RefinementWindow statement
will generate error messages.

To sweep a 2D profile along a path, use the one of the following syntax templates:
# Sweep the base element a distance along the normal to the element.
sweepElement {
base = <2D Element>
distance = <double>
}
# Sweep the base element along a vector. The vector must be
# normal to the base element.
sweepElement {
base = <2D Element>
vector = (x1,y1,z1)
}
# Sweep the base element along a polygonal path. The origin of
# the path must be normal to the base element.
sweepElement {
base = <2D Element>
path = [(x1,y1,z1)...(xn,yn,zn)]
}
# Rotate the base element about an axis parallel to the z-axis.
# The axis will be placed at the center of the base element.
sweepElement {
base = <2D Element>
angle = <double>
}
# Rotate the base element about an axis parallel to the z-axis
# which is placed at a point "p".

Sentaurus™ Mesh User Guide 79

O-2018.06
3: Doping and Refinement Examples
Creating 3D Profiles From 2D Cross Sections

sweepElement {
base = <2D Element>
point = p
angle = <double>
}
# Rotate the base element about an axis.
sweepElement {
base = <2D Element>
point = (x,y,z)
axis = (dx,dy,dz)
angle = <double>
}

In some cases, the normal of the base element is used to determine the direction of the sweep.
This normal is calculated in the following way:
■ For a polygonal base element described using the sequence [p1,p2,...,pn], the normal
is calculated as (p3 – p2)*(p1 – p2).
■ For a rectangular base element described by [p1,p2], the normal calculation is extremely
complicated and is not described here. The recommendation is to use polygonal elements,
or to swap p1 and p2 if the profile is being produced in the wrong direction.

More considerations arise when rotating an element about an axis (see Figure 5). Since there
are some degrees of freedom to perform the rotation, additional rules must be applied:
■ Only the right side of the profile is used in the sweep. This is to avoid double-defined
values, which occur when the rotation axis is contained inside the base element and the
rotation angle is more than 180° .
■ The direction of the rotation is adjusted to match the direction of the normal.

axis

normal

angle

Figure 5 Rotation of a profile about an axis; the gray portion is ignored during the sweep

80 Sentaurus™ Mesh User Guide

O-2018.06
 3: Doping and Refinement Examples
Using Particle Profiles to Specify Doping

The angle can be set to a negative value, or the orientation of the axis can be reversed, to obtain
the required profile.

When a 2D submesh profile is loaded into a 3D simulation, the default is to place it along the
xy coordinate plane. The ShiftVector and Rotation parameters must be used to place the
profile at the desired location.

For example, to place a submesh on the xz plane, passing through the point (0, 50, 50), and then
to sweep it along a path, the following can be used:
SubMesh "profile" {
Reference = "SubMesh"
Rotation {
angle = -90
axis = X
}
ShiftVector = (0 50 50)
EvaluateWindow {
Element = sweepElement {
base = rectangle [(0 45 0), (15 45 50)]
path = [(15 45 0) (15 35 0) (30 45 0) (35 25 0)]
}
DecayLength = 1
}
}

Using Particle Profiles to Specify Doping

This example illustrates the use of particle profiles for specifying the doping for a 30-nm
n-channel MOSFET. The particle information is generated by Sentaurus Process Kinetic
Monte Carlo and is stored in a TDR file named 30nm_end6.tdr. Figure 6 on page 82 shows
the results of the kinetic Monte Carlo (KMC) simulation with the gate material and gate oxide
removed. The light blue dots represent arsenic point defects, and the dark blue dots represent
boron point defects.

The command file nmos.cmd shown below is used to generate the mesh and doping for the
structure.

NOTE Before running nmos.cmd, a 3D boundary file named nmos_bnd.tdr

should exist that is consistent with the KMC particle file. If the final
generated structure is to be used in Sentaurus Device, the boundary file
must also contain the proper electrodes needed for device simulation.

Sentaurus™ Mesh User Guide 81

O-2018.06
3: Doping and Refinement Examples
Using Particle Profiles to Specify Doping

Figure 6 Discrete dopant positions generated by Sentaurus Process Kinetic Monte Carlo

The Definitions section of the command file defines the profiles and refinements that will
later be used (in the Placements section) to create the mesh and doping for the structure:
Title "Specifying doping with particle profiles"
Definitions {
Constant "PolyGateDoping" {
Species = "ArsenicActiveConcentration"
Value = 8e+19
}
Particle "BoronParticles" {
ParticleFile = "30nm_end6.tdr"
Species = "BoronActiveConcentration"
ScreeningFactor = 2.5e6
AutoScreeningFactor
Normalization
}
Particle "ArsenicParticles" {
ParticleFile = "30nm_end6.tdr"
Species = "ArsenicActiveConcentration"
ScreeningFactor = 1.0e7
AutoScreeningFactor
Normalization
}
Refinement "GlobalRefinement" {
MaxElementSize = ( .020 .020 .020 )
MinElementSize = ( .002 .002 .002 )
RefineFunction = MaxTransDiff(Variable = "DopingConcentration",Value = 1)
}
Refinement "InterfaceRefinement" {
MaxElementSize = ( .008 .004 .0002 )
MinElementSize = ( .002 .002 .0001 )

82 Sentaurus™ Mesh User Guide

O-2018.06
 3: Doping and Refinement Examples
Using Particle Profiles to Specify Doping

RefineFunction = MaxTransDiff(Variable = "SpatialCoordinates",

Value = 1e-10)
}
}

The Constant definition defines the doping that will be used for the polysilicon gate. Two
Particle profiles are used to obtain separately the boron and arsenic discrete dopants from
the KMC particle file. Separate ScreeningFactor values are specified for these two
Species. Specifying AutoScreeningFactor generally results in a smoother and more
accurate final profile in structures where there are large changes in dopant density. The
Normalization option compensates for doping loss of dopants located near the boundary.
The command file also includes a global refinement definition based on doping and an interface
refinement definition based on spatial coordinates. The latter definition is intended to force a
grid refinement using the specified element sizes.

The Placements section of the command file specifies how the profile and refinement
definitions should be used to create the structure:
Placements {
Constant "PolyGateDopingPlacement" {
Reference = "PolyGateDoping"
EvaluateWindow { Element = material ["PolySilicon"] }
}
Particle "ArsenicParticlesPlacement" {
Reference = "ArsenicParticles"
EvaluateWindow { Element=material ["Silicon"] }
}
Particle "BoronParticlesPlacement" {
Reference = "BoronParticles"
EvaluateWindow { Element=material ["Silicon"] }
}
Refinement "GlobalRefinementPlacement" {
Reference = "GlobalRefinement"
RefineWindow = material ["Silicon"]
}
Refinement "InterfaceRefinementPlacement" {
Reference = "InterfaceRefinement"
RefineWindow = Cuboid [(0.00 0.060 0.0000) (0.05 0.090 -0.0008)]
}
}

The gate material (polysilicon) is uniformly doped using the Constant profile given in the
Definitions section. The Particle profiles are only placed in the silicon portion of the
structure. The global refinement (based on doping) is also only performed in silicon. The
interface refinement is confined to the channel region of the structure and to within 8 Å of the
interface.

Sentaurus™ Mesh User Guide 83

O-2018.06
3: Doping and Refinement Examples
Generating 2D Mesh With Continuous Doping Obtained From 3D KMC File Containing Particle Information

Execution of this command file will generate a TDR file named nmos_msh.tdr. The
generated structure is shown in Figure 7.
Z
Doping Concentration [cm−3]
1.2e+21
Y
6.0e+17
2.9e+14 X
-1.6e+12
-3.7e+15
-7.6e+18

Figure 7 Mesh and doping for the structure generated by Sentaurus Mesh

Generating 2D Mesh With Continuous Doping Obtained

From 3D KMC File Containing Particle Information
This example illustrates the use of the feature that generates a continuous profile on a 2D mesh
from a 3D KMC file containing particle information. This feature allows you to evaluate the
3D doping profile and to transfer those onto a 2D mesh.

The following are the Particle subsections of the Placements section of the command file:
Placements {
Particle "BoronParticles" {
ParticleFile = "n26_final.tdr"
Species = "BoronActiveConcentration"
ScreeningFactor = 3.5e6
AutoScreeningFactor
Normalization
BoundaryExtension = 0.05
Divisions = 10
}
Particle "ArsenicParticles" {
ParticleFile = "n26_final.tdr"
Species = "ArsenicActiveConcentration"

84 Sentaurus™ Mesh User Guide

O-2018.06
 3: Doping and Refinement Examples
Performing Interface Refinement

ScreeningFactor = 1.1e7
AutoScreeningFactor
Normalization
BoundaryExtension = 0.05
Divisions = 10
}
}

The parameter boundary extension is used internally as the thickness of a 3D structure

generated by extruding a 2D structure in the z-direction. The KMC file containing the particle
information is mapped onto this 3D structure. For each 2D mesh point, a number of points
equal to the number of divisions is created, each separated by an equal amount in the z-
direction, and doping is computed on these points. The average doping is computed and is
assigned to a 2D mesh point. Figure 8 shows the input boundary and generated mesh with
particle profile.

Figure 8 Two-dimensional boundary and mesh with doping obtained from a 3D KMC file

Performing Interface Refinement

Interface refinement is specified in a similar way to the refinement on analytic functions. To
perform interface refinement, define a RefineFunction of type MaxLengthInterface and
specify the pair of materials defining the interface, the initial thickness, and a factor used to
define how this thickness should grow into the material.

Sentaurus™ Mesh User Guide 85

O-2018.06
3: Doping and Refinement Examples
Performing Interface Refinement

The following examples illustrate the use of this function:

■ This function refines silicon at the oxide interface, starting with a layer of 0.02 μm and
gradually increasing the thickness by 1.4 times:
RefineFunction = MaxLenInt(Interface("Silicon","Oxide"), Value = 0.02,
Factor = 1.4)
■ This function refines all interfaces, creating a single layer of 0.01 μm :
RefineFunction = MaxLengthInterface(Interface("All","All"), Value=0.01)
■ This function refines all contacts in the mesh:
RefineFunction = MaxLenInt(Interface("All","Contact"), Value=0.01)
■ This function refines around a single contact:
RefineFunction = MaxLengthInterface(Interface("All","Gate"), Value=0.01,
UseRegionNames)

By default, the interface refinement is performed only on the first material of the specified pair
of materials describing the interface. To refine on both sides of the interface, use the
DoubleSide keyword:
RefineFunction = MaxLenInt(Interface("Silicon","Contact"), Value=0.01,
DoubleSide)

Ignoring Interfaces Between Regions of the Same Material

When Sentaurus Mesh performs refinement across interfaces, it internally splits the edges
crossing the interfaces into segments that are contained completely inside each region. Then,
it proceeds to analyze the refinement criteria on each segment independently. In some
applications, the doping concentration on each region is constant, so no refinement is applied
since the gradient along each segment is zero.

However, sometimes, you may want to define different doping concentrations on adjacent
regions of the same material and may want the code to ignore the interface between those
regions so that Sentaurus Mesh can refine across the interface. In this case, specify the
parameter skipSameMaterialInterfaces in the AxisAligned section of the command
file to obtain the required effect.

86 Sentaurus™ Mesh User Guide

O-2018.06
 3: Doping and Refinement Examples
Offsetting Mesh Generation

Offsetting Mesh Generation

This section presents examples that illustrate using the offsetting mesh generator.

Simple Example
This example shows all the relevant parameters in the Offsetting section that are supported
by Sentaurus Mesh. The input structure is shown in Figure 9.

Figure 9 Simple structure

Title ""
IOControls {
EnableSections
}
Definitions {
Refinement "R5" {
MaxElementSize = ( 0.026 0.026 0.026 )
}
}
Placements {
Refinement "GDJ_RP" {
Reference = "R5"
RefineWindow = Cuboid [(-0.2 -0.2 0) (0.20 0.2 0.5)]
}
}

Sentaurus™ Mesh User Guide 87

O-2018.06
3: Doping and Refinement Examples
Offsetting Mesh Generation

Offsetting {
noffset {
hlocal=0
}
noffset material "Silicon" {
maxlevel = 5
}
noffset material "Oxide" {
maxlevel = 5
}
noffset material "Silicon" "Oxide" {
hlocal=0.002
factor=1.5
}
noffset material "Oxide" "Silicon" {
hlocal=0.002
factor=1.5
}
noffset region "RTrench" "RBulk" {
hlocal=0.002
factor=1.5
}
noffset region "RBulk" "RTrench" {
hlocal=0.002
factor=1.5
}
}

In this command file, the global hlocal value is set to zero. Later, a nonzero hlocal value
and factor are set using the material interface section. The default maxlevel value of 200 is
also reset to 5 using a material section.

If the command file only includes the Offsetting section, without specifying any refinement
criteria in the Definitions and Placements sections, the resultant mesh will be coarse as
shown in Figure 10 on page 89.

88 Sentaurus™ Mesh User Guide

O-2018.06
 3: Doping and Refinement Examples
Offsetting Mesh Generation

Figure 10 Coarse mesh generated with layers and without refinement criteria

With the refinement specified in the command file, the generated mesh is shown in Figure 11.

Figure 11 Mesh generated with layers along with refinement criteria

Sentaurus™ Mesh User Guide 89

O-2018.06
3: Doping and Refinement Examples
Offsetting Mesh Generation

Layering From All Boundaries

Title ""
IOControls {
EnableSections
}
Definitions {
Refinement "RefinementDefinition_1" {
MaxElementSize = ( 0.2 0.2 0.2 )
MinElementSize = ( 0.001 0.001 0.001 )
}
}
Placements {
Refinement "RefinementPlacement_1" {
Reference = "RefinementDefinition_1"
RefineWindow = region ["R_Silicon"]
}
Refinement "RefinementPlacement_2" {
Reference = "RefinementDefinition_1"
RefineWindow = region ["R_Oxide"]
}
}
Offsetting {
noffset {
hlocal=0.01
maxlevel=6
}
noffset material "Oxide" "Silicon" {
factor=2
}
}

In this command file, the global parameter maxlevel is set to 6 and the global hlocal is a
nonzero value. This results in layering only from interfaces excluding exterior boundaries. If
layering is required from all interfaces including boundaries that do not share any, the
Offsetting section can be modified as shown:
Offsetting {
noffset {
maxlevel=6
}
noffset material "All" "All" {
hlocal=0.01
factor=2
}
}

90 Sentaurus™ Mesh User Guide

O-2018.06
 3: Doping and Refinement Examples
Localizing the Refinement Using Cuts

The string "All" refers to all materials in the input structure. The resultant meshes that are
generated using the original Offsetting section and the modified Offsetting section are
shown in Figure 12.

Figure 12 (Left) Mesh with layering from all interfaces and from boundaries with nonzero
global hlocal value and (right) mesh generated with modified Offsetting section
having layering only at interfaces

Localizing the Refinement Using Cuts

By design, the axis-aligned algorithm always creates the mesh by refining an initial box, which
contains the whole device. This creates problems when the external shape of the device must
be modified, because this will change the bounding box of the device, thereby altering the
location of the mesh nodes.

To help resolve this situation, the axis-aligned algorithm offers the possibility of defining an
initial array of boxes from which to start the refinement. This is performed through the xCuts,
yCuts, and zCuts parameters in the AxisAligned section.

Each cut defines and initial refinement line that runs throughout the whole device. The boxes
created with these initial lines can be refined independently of each other. Therefore, if the
shape of the device changes and the cuts are adjusted accordingly, the mesh should stay the
same in the sections where the boxes have remained unchanged.

Sentaurus™ Mesh User Guide 91

O-2018.06
3: Doping and Refinement Examples
Localizing the Refinement Using Cuts

The following example uses two lines to generate a total of three initial boxes. The lines are
placed on either side of the channel and can be used to parameterize a set of structures where
the only difference is the channel length:
AxisAligned {
xCuts = (4.837 7.156)
}

Figure 13 displays the channel of an NMOS structure that has been refined using the standard
refinement parameters.

-1.5

-1

-0.5
Y [µm]

Doping
Concentration [cm–3]
0
1.6e+20
3.8e+17
9.1e+14
0.5
2.1e+12
-4.7e+13
-2.0e+16

4.2 4.4 4.6 4.8 5 5.2 5.4 5.6 5.8 6 6.2 6.4 6.6 6.8 7 7.2 7.4 7.6 7.8
X [µm]
Figure 13 NMOS structure refined using standard refinement

When the xCuts parameter is used, two refinement lines are placed at locations x=4.837 and
x=7.156 (see Figure 14 on page 93). If the channel length is increased by 0.01 μm , the second
x-line could be placed at 7.157, thereby preserving the mesh on either side of the channel.

92 Sentaurus™ Mesh User Guide

O-2018.06
 3: Doping and Refinement Examples
Localizing the Refinement Using Cuts

-1.5

-1

-0.5
Y [µm]

Doping
Concentration [cm–3]
0
1.6e+20
3.8e+17
9.1e+14
0.5
2.1e+12
-4.7e+13
-2.0e+16

4.2 4.4 4.6 4.8 5 5.2 5.4 5.6 5.8 6 6.2 6.4 6.6 6.8 7 7.2 7.4 7.6 7.8
X [µm]

Figure 14 NMOS structure refined using the xCuts option

Another possibility is to use the fitInterfaces parameter in the AxisAligned section of

the command file. This parameter works best on simple devices where all interfaces are axis
aligned. Figure 15 shows that the device can be overrefined when the interfaces are not simple.

-1.5

-1

-0.5
Y [µm]

Doping
Concentration [cm–3]
0
1.6e+20
3.7e+17
8.9e+14
0.5
2.0e+12
-5.0e+13
-2.1e+16
4.2 4.4 4.6 4.8 5 5.2 5.4 5.6 5.8 6 6.2 6.4 6.6 6.8 7 7.2 7.4 7.6 7.8
X [µm]

Figure 15 NMOS structure refined using the fitInterfaces parameter

Sentaurus™ Mesh User Guide 93

O-2018.06
3: Doping and Refinement Examples
Using Analytic Functions for Refinement I

Using Analytic Functions for Refinement I

Figure 16 illustrates the use of general analytic functions to specify profiles.

The function 0.1 sin ( x ) sin ( y ) is used as a profile and linear interpolation
("ElectrostaticPotential") is used to compute the required local element size. The
following command file segment illustrates the syntax:
Definitions {
Refinement "Region_1" {
MaxElementSize = (1 1)
MinElementSize = (0.01 0.01)
RefineFunction = MaxTransDiff(Variable = "ElectrostaticPotential",
Value = 0.01)
}
AnalyticalProfile "Profile_1" {
Species = "ElectrostaticPotential"
Function = General(init="a=0.1",function = "a*sin(x)*sin(y)",value = 0)
}
}
Placements {
Refinement "Region_1" {
Reference = "Region_1"
}
AnalyticalProfile "Profile_1" {
Reference = "Profile_1"
EvaluateWindow {
Element = rectangle [(0 0), (9.43 9.43)]
}
}
}

Analytic function

Generated mesh

Figure 16 Use of analytic refinement functions

94 Sentaurus™ Mesh User Guide

O-2018.06
 3: Doping and Refinement Examples
Using Analytic Functions for Refinement II

Using Analytic Functions for Refinement II

This example illustrates the use of a general analytic function to prescribe 3D refinement based
on a 3D analytic function. The domain is a cube. Figure 17 shows the generated mesh.

Figure 17 Use of analytic refinement functions

Definitions {
Refinement "Region_1" {
MaxElementSize = (4 4 4)
MinElementSize = (0.01 0.01 0.01)
RefineFunction = MaxTransDiff(Variable = "ElectrostaticPotential",
Value = 10000.0)
}
AnalyticalProfile "Profile_1" {
Species = "ElectrostaticPotential"
Function = General(init="a=0.1",function = "a*x*x*y*y*z*z",value = 0)
}
}
Placements {
Refinement "Region_1" {
Reference = "Region_1"
}
AnalyticalProfile "Profile_1" {
Reference = "Profile_1"
EvaluateWindow {
Element = cuboid [ ( 0 0 0 ), ( 10 10 10 )]
}
}
}

Sentaurus™ Mesh User Guide 95

O-2018.06
3: Doping and Refinement Examples
Using Analytic Functions for Refinement II

96 Sentaurus™ Mesh User Guide

O-2018.06
CHAPTER 4 Tensor-Product Examples

This chapter uses various examples to demonstrate the applications

of the Tensor section of the command file.

Simple Cube
This example illustrates the effectiveness of various features such as maxBndCellSize,
maxCellSize, refinement using window, and grading. The following is the command file
used to generate the tensor-product mesh:
Tensor {
Mesh {
maxBndCellSize direction "x" 0.001
maxCellSize region "Region_0" 0.1
window "testbox" 0.8 1.2 0.8 1.2 0.8 1.2
minNumberOfCells window "testbox" 20
grading = { 1.1 1.1 1.1 }
}
}

Figure 18 shows the geometry of this example. In the above command file, the first parameter
is maxBndCellSize, which is constrained in the x-direction by specifying the direction
option. As a result, clustering is obtained only near the boundaries that are normal to the x-axis
(see Figure 18, right).

Figure 18 Geometry of a cube with a width of 2.0 units in each direction: (left) the input
geometry and (right) corresponding tensor mesh

Sentaurus™ Mesh User Guide 97

O-2018.06
4: Tensor-Product Examples
Simple Cube

The second parameter maxCellSize is specified within a region. Since no direction

option is used in this parameter, the mesh generator will try to obtain the same cell size in all
three directions. To obtain refinement in the center of the geometry, the third parameter
window is defined. The required refinement within a window can be obtained by specifying
either maxCellSize or minNumberOfCells.

In this example, minNumberOfCells is used to specify the refinement. As a result of this

refinement parameter, clustering of lines is visible in Figure 18 on page 97. The grading
parameter is also defined in this command file. This is used to obtain a smooth variation of the
cell sizes between various cell sizes (see Figure 19).

In Figure 18, the clustering of cells normal to the x-axis is visible as per the specification of the
minimum boundary cell size parameter in the command file. The refinement in the center of
the cube is due to the specification of the minNumberOfCells parameter in the command file.
This refinement is constrained to a "testbox" window.

Figure 19 Smooth variation of cell size from minimum to maximum cell size according to
specified grading factor

98 Sentaurus™ Mesh User Guide

O-2018.06
 4: Tensor-Product Examples
Using Boundary and Command Files to Generate Doping and Refinement

Using Boundary and Command Files to Generate Doping

and Refinement
This example shows how to use the tensor-product mesh generator to represent an
approximation of the actual regions defined in a boundary file. The geometry is shown in
Figure 20. In this case, the default mesh generation parameters are used. A doping refinement
section is provided in the command file.

Figure 20 Geometry of an input file

Figure 21 shows the tensor mesh and the corresponding doping data. Note that the axis-aligned
interfaces are represented accurately by the tensor mesh. However, if the geometry has a curved
region, the resulting mesh will be an approximation of the boundary, as shown in Figure 22 on
page 100.

Doping Concentration [cm−3]

1.1e+20
1.2e+17
1.2e+14
-1.8e+12
-1.8e+15
-1.7e+18

Figure 21 (Left) Tensor mesh and (right) the doping data interpolated onto tensor mesh

Sentaurus™ Mesh User Guide 99

O-2018.06
4: Tensor-Product Examples
Thin Regions

Figure 22 (Left) Actual curve region in the geometry and (right) corresponding
approximation of this region in tensor mesh

Thin Regions
This example shows how insufficient cell resolution in a region will result in elements that are
one dimension less than the model dimension. The geometry contains a thin aluminum region
shown in Figure 23. Since the local cell size is not small enough to resolve this aluminum
region, this region is represented as one-dimensional elements in the output as shown in
Figure 24 on page 101.

Figure 23 (Left) Geometry of the example and (right) detail of thin aluminum region

In Figure 24, the aluminum region is not resolved properly as the locally cell size is larger than
the aluminum region. As a result, this region contains two faces and rest of the region is
represented as a set of edges connecting the two faces. Similarly, in three-dimensional models,
unresolved regions and contacts will be written as thin surface sheets. You will be given a
warning indicating the presence of completely or partially unresolved regions.

100 Sentaurus™ Mesh User Guide

O-2018.06
 4: Tensor-Product Examples
Computing Cell Size Automatically (EMW Applications)

The following is the command file used to generate the tensor-product mesh:
Tensor {
Mesh {
maxCellSize material "Oxide" 0.08
maxCellSize material "Silicon" 0.08
grading = { 1.1 1.1 1.1 }
}
}

Figure 24 (Left) Corresponding tensor mesh and (right) detail of tensor mesh near
aluminum region

Computing Cell Size Automatically (EMW Applications)

This example shows the command file structure that is used to compute cell sizes automatically
using the tensor mesh generator. The tensor mesh generator reads the Mesh section of the
command file first and then reads the EMW section. Since the cell size for the material
Insulator1 is defined in this section, it is not recomputed. The cell sizes for the rest of the
materials are computed using the optical database table, which is defined in either the user-
defined Sentaurus Device parameter file or the default Sentaurus Device parameter file.

Sentaurus™ Mesh User Guide 101

O-2018.06
4: Tensor-Product Examples
Computing Cell Size Automatically (EMW Applications)

The computed cell size is a function of the:

■ Wavelength
■ Nodes per wavelength
■ Norm of a complex refractive index

In this example, the wavelength is equal to 2.0 μm and the nodes per wavelength is 5. The cell
size is computed for all materials except Insulator since its cell size is specified in the Mesh
section.

Among the materials present in the shown structure (see Figure 25), the computed cell size is
smallest for Aluminum. Because of this, the mesh lines are more clustered in the corresponding
region:
Tensor {
Mesh {
maxCellSize material "Insulator1" 0.1
}
EMW {
wavelength = 2.0
npw = 5
}
}

Figure 25 (Left) Boundary and (right) corresponding tensor mesh generated using the
EnableEMW option in the IOControls section of the command file

102 Sentaurus™ Mesh User Guide

O-2018.06
CHAPTER 5 Tools Section

This chapter illustrates the use of the Tools section of the command
file for Sentaurus Mesh.

Activating the Tools Section

To activate the Tools section in the command file of Sentaurus Mesh, specify either the
EnableSections or the EnableTools option in the IOControls section of the command
file (see IOControls Section on page 6).

Reflecting and Sweeping Mesh

In this example, a two-dimensional (2D) structure is taken and reflected. A three-dimensional
(3D) mesh is then generated by sweeping the reflected mesh. The command file is:
Tools {
Reflection {
axis = xmin
map "R.PolyReox" = "R.PolyReox_new"
}
Sweepmesh {
extension = 0.1
steps = 5
}
}

The input mesh is shown in Figure 26 on page 104. The first step reflects the mesh structure
about the xmin coordinate. The map statement renames the mirrored region R.PolyReox to
R.PolyReox_new.

Sentaurus™ Mesh User Guide 103

O-2018.06
5: Tools Section
Reflecting and Sweeping Mesh

DopingConcentration [cm−3]
1.1e+20
1.2e+17
1.2e+14
-1.8e+12
-1.8e+15
-1.6e+18

Figure 26 Input structure

The reflected mesh is shown in Figure 27. Since Sweepmesh is specified in the command file,
the reflected mesh becomes input to this tool.

DopingConcentration [cm−3]
1.1e+20
1.2e+17
1.2e+14
-1.8e+12
-1.8e+15
-1.6e+18

Figure 27 Reflected mesh

A 3D mesh is generated by sweeping the reflected mesh in the z-direction by 0.1 μm . Then,
this mesh is divided into five sections along the z-direction as shown in Figure 28.

Doping Concentration [cm−3]

1.1e+20
1.2e+17
1.2e+14
-1.8e+12
-1.8e+15
-1.6e+18

Figure 28 Mesh after being swept in the z-direction

104 Sentaurus™ Mesh User Guide

O-2018.06
 5: Tools Section
Slicing a 3D Mesh Using a Plane and Its Location

Slicing a 3D Mesh Using a Plane and Its Location

In this example, a 3D mesh is sliced to obtain a 2D mesh. The command file is:
Tools {
Slice {
normal = (0 1 0)
location = (0 0.0075 0)
}
}

The input mesh is shown in Figure 29. The mesh is sliced with the y-plane placed at the
specified location (0 0.0075 0). This results in a 2D mesh slice shown in Figure 30 on
page 106.

Figure 29 The input 3D mesh to be sliced

Sentaurus™ Mesh User Guide 105

O-2018.06
5: Tools Section
Cutting a 3D Mesh

Figure 30 The 2D mesh slice generated from the mesh shown in Figure 29 on page 105

Cutting a 3D Mesh
In this example, a cube mesh is taken as an input and three cutting planes are used to create a
wedge. The command file used in this example is:
Tools {
Cut {
normal = (0 0 1)
location = (0.05 0.05 0.05)
}
Cut {
normal = (0 -0.70711 0.70711)
location = (0.05 0.05 0.05)
}
Cut {
normal = (0 1 0)
location = (0.05 0.05 0.05)
}
}

106 Sentaurus™ Mesh User Guide

O-2018.06
 5: Tools Section
Converting a Tetrahedral Mesh to a Hybrid Mesh

In this code block, three cutting planes are specified. The input mesh is processed with the first
cutting plane, and its output is given as an input to the next cutting plane. The input mesh and
the final output mesh are shown in Figure 31.

Figure 31 (Left) Input mesh and (right) final wedge created by using three cutting planes

Converting a Tetrahedral Mesh to a Hybrid Mesh

The following example translates a mesh and converts it to a hybrid mesh. The command file
used in this example is:
Tools {
Set Transformation {
translation = ( 3 7 1 )
}
Apply Transformation
Mesh2Hybrid
}

The log file contains information about a number of different element types in the converted
mesh. In this example, approximately 55% of element reduction is achieved when compared
to the input mesh.

Sentaurus™ Mesh User Guide 107

O-2018.06
5: Tools Section
Generating Randomized Doping From Continuous Doping

Figure 32 (Left) Tetrahedral mesh and (right) corresponding hybrid element produced by
Mesh2Hybrid algorithm

Generating Randomized Doping From Continuous Doping

The following example illustrates a utility that randomizes a continuous doping. A mesh with
continuous doping, along with a command file, is given as input. The command file for this
example is:
Title "DopingRandomizer Example"
# Use "snmesh nmos00.cmd" to run. Assumes nmos00.tdr exists.
IOControls {
EnableSections
}
Tools {
RandomizeDoping {
DopingAssignment = "Sano"
# ContinuousContactDoping
NumberOfRandomizedProfiles = 1
FileIndex = 1
Material "Silicon" {
Species "BoronActiveConcentration" {
ScreeningFactor = 2.5e6
AutoScreeningFactor
}
Species "ArsenicActiveConcentration" {
ScreeningFactor = 1.3e7
AutoScreeningFactor
}
}
}
}

108 Sentaurus™ Mesh User Guide

O-2018.06
 5: Tools Section
Generating Randomized Doping From Continuous Doping

The method chosen in this example is "Sano". The number of randomized profiles is 1, and
FileIndex is set to 1. Since only the material "Silicon" is specified, only silicon regions
will be randomized. Other materials will retain their original continuous doping. Figure 33
shows the input structure mesh with continuous doping.

Figure 33 Input mesh with continuous doping

Figure 34 (left) shows the randomized doping generated using DopingAssignment="Sano".

Later, the command file is modified to use the other available methods. The middle structure
in Figure 34 is generated using the NGP method, and third structure is generated using the CIC
method.

Figure 34 Meshes with randomized doping generated using the (left) Sano method,
(middle) NGP method, and (right) CIC method

Sentaurus™ Mesh User Guide 109

O-2018.06
5: Tools Section
Slicing a 3D Mesh Using a Segment and a Direction

Slicing a 3D Mesh Using a Segment and a Direction

In this example, a 3D mesh is sliced to obtain a 2D mesh using a segment feature. The
command file is:
Tools {
Slice {
Direction = Z
Startpoint = (0.15 0.0)
Endpoint = (0.15 0.75)
}
}

In the above Tools section, a segment with a starting point and an endpoint on a constant Z-
plane is specified. With this information, a bounding box of the input structure (shown in
Figure 35 (left)) is computed and used in the construction of a plane defined by the bounding
box coordinates of [(0.15 0.0 zmin) (0.15 0.75 zmax)]. The input boundary is sliced
with this plane and the result is shown in Figure 35 (right). This boundary slice file also
contains transformation information that can be used to place this slice back into 3D space for
later applications.

Figure 35 (Left) Input boundary and (right) 2D slice generated using slice utility

110 Sentaurus™ Mesh User Guide

O-2018.06
 5: Tools Section
Creating Profiles in an Existing Mesh

Creating Profiles in an Existing Mesh

This example shows the usage of the creating profiles utility. The command file contains
information about an existing mesh and a mesh command file containing profile information.
The profiles specified in the command file are created in the input mesh without changing the
mesh itself. The command file for this example is:
Tools {
CreateProfiles {
SrcMesh = "n6_0_msh.tdr"
CmdFile = "n6_msh.cmd"
}
}

In the above CreateProfiles section, the source mesh file (shown in Figure 36 (left)) and
the mesh command file are specified.

BoronActive BoronActive
Concentration [cm−3] Concentration [cm−3]
3.0e+18 1.0e+20
9.6e+17
3.1e+17
9.8e+16
3.1e+16
1.0e+16

Figure 36 (Left) Input mesh showing BoronActiveConcentration profile before update and
(right) mesh with updated BoronActiveConcentration

The mesh command file contains the following information related to

BoronActiveConcentration:
Title "Untitled"
Definitions {
Constant "substrateDop" {
Species = "BoronActiveConcentration"
Value = 1e+20
}
}
Placements {
Constant "substrateDop" {
Reference = "substrateDop"
EvaluateWindow { Element = region ["substrate"] }
}
}

Sentaurus™ Mesh User Guide 111

O-2018.06
5: Tools Section
Stretching a Mesh

The BoronActiveConcentration profile in the input mesh is recreated without changing

other profiles and, accordingly, the doping concentration is updated. The output mesh file is
shown in Figure 36 on page 111 (right).

Stretching a Mesh
This example shows the usage of the stretch utility. The command file contains information
about the location of the starting point of the stretch, the direction of the stretch, and the length
of the stretch. The command file for this example is:
Tools {
Stretch {
location = (0.12 0.005 0)
direction = X
length = 0.05
}
}

The input mesh is shown in Figure 37 (left). With the information in the command file, a new
column of elements is added at the specified location. The output stretch mesh is shown in
Figure 37 (right). After stretching, the length of the mesh in the specified direction is increased
by a specified length. The unit of length is the same as the input mesh.

Figure 37 (Left) Input mesh before stretch and (right) mesh after stretch

112 Sentaurus™ Mesh User Guide

O-2018.06
CHAPTER 6 Delaunization Algorithm

This chapter describes the delaunization algorithm used by

Sentaurus Mesh.

Overview
A delaunization algorithm is available for 3D models in Sentaurus Mesh. This algorithm is
based on a conforming Delaunay triangulation–type of algorithm, but it is more stable,
generating meshes for complex structures with sharp input angles that could not previously be
handled. This delaunizer also produces fewer mesh nodes than the previous algorithm.
Reference [1] is an excellent book on Delaunay mesh generation.

The algorithm uses two independent structures to generate the final mesh: a set of surface faces
(for example, the input boundary and some isosurfaces, or rectangular faces originating from
user-defined refinement inside Sentaurus Mesh) and a background three-dimensional generic
Delaunay triangulation.

The algorithm works in the following way:

■ Ridges and corners are classified.
■ A set of protection spheres is generated around ridges and corners.
■ A 2D surface delaunization algorithm is applied. This algorithm flips all nonridge edges
that do not meet the Delaunay criterion.
■ Each ridge that does not meet the Delaunay criterion is refined (see [2]).
■ Each surface face that does not meet the Delaunay criterion is refined (see [2]).
■ Each element that does not meet the quality criteria is refined.
■ The surface faces that have not been recovered by refinement are recovered using a
constrained Delaunay triangulation (CDT) algorithm (see [3][4]).
■ Slivers are removed (see [5][6]).
■ A material is assigned to each tetrahedron in the final triangulation.

Sentaurus™ Mesh User Guide 113

O-2018.06
6: Delaunization Algorithm
Generating Ridges and Corners

Generating Ridges and Corners

During the first stage, the algorithm detects the faces that are coplanar. Every edge that bounds
a coplanar set of faces is labeled a ridge. Every point that connects two non-collinear ridges is
labeled a corner.

By default, two faces are coplanar if the angle between them is less than coplanarityAngle
and the surface deformation that can result from flipping the common edge is less than
coplanarityDistance.

Protecting Ridges and Corners

In general, conforming Delaunay algorithms do not perform well if the input contains sharp
angles between adjacent faces on the surface (in general, of less than 60° ). A generic algorithm
would refine excessively around ridges and corners that define a very sharp angle.
Occasionally, generic algorithms do not stop and the algorithms collapse.

The algorithm for ridge and corner refinement carefully refines around sharp corners and
ridges, defining a set of spheres that protect these entities. Refinement points that are inside
these spheres are snapped to the surface of the sphere. This produces constructions that
resemble isosceles triangles, which are well suited to Delaunay-type algorithms because
isosceles triangles contain their circumscribed centers inside them.

Conforming Delaunay Triangulation Algorithm

The conforming Delaunay triangulation (CDT) algorithm enables the delaunizer to produce
meshes that are near-Delaunay after relaxing the Delaunay criterion.

After the faces have been refined to meet the (possibly relaxed) Delaunay criterion, some
surface faces may be missing from the background 3D Delaunay mesh. The CDT algorithm
inserts those faces into the background triangulation using a sequence of 3D face flips.

This algorithm is very complex, therefore, you can expect long runtimes if the Delaunay
criterion is relaxed too much at locations with many faces to be recovered (such as locations
with a lot of refinement in Sentaurus Mesh).

114 Sentaurus™ Mesh User Guide

O-2018.06
 6: Delaunization Algorithm
Optimizing Elements

Optimizing Elements
After the CDT algorithm is finished, the quality of the elements in the mesh may not be
optimal. Therefore, the algorithm performs an extra refinement step, which eliminates all
elements that do not meet the quality criteria specified by users. Two quality criteria are
available:
■ The maximum solid angle inside an element.
■ The maximum ratio between the circumscribed spheres of neighboring elements.

Any element that does not meet the quality criteria will be refined. The algorithm used to refine
the elements is based on the Delaunay refinement technique, which inserts a node at the
Voronoï center of the element and updates the neighboring triangulation. If the Voronoï center
of the element lies too close to the surface, the surface will be refined.

Eliminating Slivers
The last step in the delaunization involves the elimination of sliver elements. To perform this,
the algorithm uses a variation of the sliver exudation technique. This technique assigns weights
to the nodes in the triangulation and uses them to compute a weighted Delaunay triangulation.
The weights are increased selectively to eliminate slivers locally in the triangulation.

The sliver elimination step changes the regular Voronoï diagram, producing Voronoï cells that
have negative sides.

The amount of damage is proportional to the weight applied to the mesh nodes. Therefore, the
algorithm includes a parameter, called sliverDistance, to control the amount of damage to
the mesh. This parameter represents the maximum weight applied to a mesh node.

References
[1] S.-W. Cheng, T. K. Dey, and J. R. Shewchuk, Delaunay Mesh Generation, Boca Raton,
Florida: CRC Press, 2013.
[2] J. R. Shewchuk, “Mesh Generation for Domains with Small Angles,” in 16th Annual
Symposium on Computational Geometry, Hong Kong, pp. 1–10, June 2000.
[3] J. R. Shewchuk, “Constrained Delaunay Tetrahedralizations and Provably Good
Boundary Recovery,” in Proceedings of the 11th International Meshing Roundtable,
Ithaca, NY, USA, pp. 193–204, September 2002.

Sentaurus™ Mesh User Guide 115

O-2018.06
6: Delaunization Algorithm
References

[4] J. R. Shewchuk, “Updating and Constructing Constrained Delaunay and Constrained

Regular Triangulations by Flips,” in 19th Annual Symposium on Computational
Geometry, San Diego, CA, USA, pp. 181–190, June 2003.
[5] S.-W. Cheng et al., “Sliver Exudation,” Journal of the ACM, vol. 47, no. 5, pp. 883–904,
2000.
[6] H. Edelsbrunner and D. Guoy, “An Experimental Study of Sliver Exudation,” in
Proceedings of the 10th International Meshing Roundtable, Newport Beach, CA, USA,
pp. 307–316, 2001.

116 Sentaurus™ Mesh User Guide

O-2018.06
APPENDIX A Formulas for Analytic Profiles

Sentaurus Mesh implements a complete set of analytic models to

describe a wide range of different situations. The reason for
implementing analytic profiles is to have a flexible tool to substitute
process simulation results efficiently and within a reasonable time.

This appendix discusses:

■ General concepts
■ Models that are available along the primary direction
■ Models that are available along the lateral direction

Although the formulas are designed according to the models associated with impurity
concentrations, analytic profiles can be used for any type of variable defined in the output files.

General Concepts
The impurity concentrations can be represented by a set of 1D, 2D, and 3D analytic models.
To describe each analytic model, two main directions must be defined: the primary direction
that is perpendicular to the reference region and the lateral direction that is parallel to the
reference region.

Along each direction, one function is defined, that is, the primary function and lateral function.
The correct combination of both functions allows you to have an analytic description of a
species concentration.

Local Coordinate Systems, Valid Domains, and Reference

Regions
The valid domain for the analytic models depends on the reference region, which is defined
using a dimension-dependent geometric element, and it is placed along the lateral direction. By
combining the reference region and primary direction, it is possible to define a local coordinate
system for each analytic function.

Sentaurus™ Mesh User Guide 117

O-2018.06
A: Formulas for Analytic Profiles
General Concepts

One-Dimensional Profiles

One-dimensional profiles require only the definition of the primary function, which is applied
along the x-axis. The primary direction and valid domain are defined using a vector. The
reference region for a profile is defined by using a geometric element, that is, a point. Figure 38
shows the scheme used for the 1D case.
Reference Point

Valid Domain and Direction

Figure 38 Primary direction in one dimension

Two-Dimensional Profiles

For 2D profiles, the reference region is defined using a baseline. The primary direction is the
normal vector to the baseline and the lateral direction is parallel to the baseline. Figure 39
shows the general scheme of the local coordinate system and the valid domain. The valid
domain for both the primary and lateral functions is defined by sweeping the primary direction
vector along the lateral direction.

Reference Line or
Baseline
Lateral Direction

Primary Direction

Lateral Direction
Primary Domain

Lateral Domain

Figure 39 Primary and lateral directions in two dimensions

118 Sentaurus™ Mesh User Guide

O-2018.06
 A: Formulas for Analytic Profiles
General Concepts

Three-Dimensional Profiles

For 3D profiles, the reference region is defined using a surface. The primary direction is the
normal vector to the surface and the lateral direction is the plane perpendicular to the primary
direction. Figure 40 shows the general scheme of the local coordinate system and the valid
domain. The valid domain for both primary and lateral functions is defined by sweeping the
primary direction vector along the surface.

Lateral Direction
Reference Surface

Lateral Direction

Primary Direction

Figure 40 Primary and lateral directions in three dimensions

General Implantation Models

In general, impurity concentrations can be expressed as:

doping(x p, x l) = g  x p ⋅ f  x l (3)

where:
■ g  x p represents the primary function in the local coordinate system.
■ f  x l represents the lateral function in the local coordinate system.

The most important functions used as models are Gaussian functions and error functions. For
the remainder of this appendix, functions along the primary direction are referred to as g  y
and functions along the lateral direction, as f  x . The indices y and x are important to
distinguish parameters among the different directions.

Each model is defined by the minimum set of parameters. This section presents a basic
formulation of each model by using the minimum set of parameters. Subsequent sections show
how to obtain this minimum set from different input or initial conditions.

Sentaurus™ Mesh User Guide 119

O-2018.06
A: Formulas for Analytic Profiles
General Concepts

Gaussian Function

The minimum set of parameters to define a Gaussian function is:

–3
■ Peak concentration ( C peak ) [ cm ]
■ Peak position ( y peak ) [ μm ]
■ Length ( GLength y ) [ μm ] or standard deviation ( stdDev y ) [ μm ]

Using these parameters, the Gaussian is defined by:

1 y – y peak 2 y – y peak 2
g ( y ) = C peak ⋅ exp  – --- ⋅ --------------------  = C peak ⋅ exp  – -----------------------  (4)
 2 stdDev y   GLength y 

Figure 41 shows the model schematically.

Concentration (y) [cm-3]
Cpeak 10

Cpeak * exp(-1) 4

GLengthy
2

y value [µm]
0 2 4 6 8 10
yPeak

Figure 41 General shape of Gaussian functions

Error Function

The minimum set of parameters to define an error function as a doping profile is:
–3
■ Maximum concentration ( C max ) [ cm ]
■ Symmetry position ( y sym ) [ μm ]
■ Length ( ELength y ) [ μm ]

Using these parameters, the error function is defined by:

C max y sym – y C max y – y sym
g ( y ) = ------------ ⋅  1 + erf ----------------------  = ------------ ⋅  1 – erf ----------------------  (5)
2  ELength y  2  ELength y 

120 Sentaurus™ Mesh User Guide

O-2018.06
 A: Formulas for Analytic Profiles
General Concepts

The function is symmetric with respect to the inflection point. Figure 42 shows the feature.

Concentration (y) [cm−3]

Cmax 10

6
Cmax/2

2
ELengthy
Cmax/2(1−erf(1))
y value [µm]
0 2 4 6 8 10
ysym

Figure 42 General shape of error functions

Other Relevant Parameters

To have flexible models, some special parameters must be considered. These are not included
in the standard formulation. However, by applying some definitions, the basic set can be
obtained from them.

Dose

From a process simulation perspective, implantation functions are determined giving the dose
concentration of the profiles. The peak concentration value can be obtained from the Dose (see
–2
Available Models Along the Primary Direction on page 122). Dose is given in atoms per cm .

The general definition of Dose is:

∞
Dose = 0 g ( y ) dy (6)

For Gaussian functions, the Dose is represented as:

∞
y – y peak 2
 Cpeak ⋅ exp  – --2- ⋅ --------------------  dy
1
Dose = (7)
stdDev y 
0

Sentaurus™ Mesh User Guide 121

O-2018.06
A: Formulas for Analytic Profiles
Available Models Along the Primary Direction

C peak ⋅ π ⋅ stdDev y  y peak 

Dose = ------------------------------------------------ ⋅  1 + erf -------------------------------  (8)
2  2 ⋅ stdDev y 

For error functions, the Dose is defined as:

∞
C max y sym – y
Dose = - ⋅  1 + erf
 ----------- ----------------------  dy (9)
2  ELength y 
0

C max ⋅ ELength y y sym y sym y sym 2

Dose = --------------------------------------- ⋅  ---------------------- ⋅ 1 + erf  ---------------------- + ------- ⋅ exp –  ---------------------- 
1
 ELength y  ELength y  ELength y
(10)
2 π

Values at the Junction

Junction Concentration and Depth are parameters used to define either Gaussian or error
functions. A complete description of these parameters and how they can replace the standard
deviation in the basic formulation is explained in Available Models Along the Primary
Direction.

Length

For Gaussian functions, GLength represents the distance between the peak position and a place
where the concentration decays by a factor of exp(–1) (36%) with respect to the peak
concentration (see Figure 41 on page 120). The relationship between the length and standard
deviation for Gaussian functions is:

GLength y = 2 ⋅ stdDev y (11)

Available Models Along the Primary Direction

The following models in Sentaurus Mesh are applied along the primary direction:
■ Gaussian functions
■ Error functions
■ Constant functions
■ 1D external profiles

122 Sentaurus™ Mesh User Guide

O-2018.06
 A: Formulas for Analytic Profiles
Available Models Along the Primary Direction

Gaussian Functions
The basic set for Gaussian functions is formed by C peak , y peak , and stdDev y . According to user
input, the basic set of parameters can be specified in different ways depending on the
parameters used to calculate C peak and stdDev y :
■ Peak Concentration and Standard Deviation
The basic set is complete (see Eq. 4) and no basic parameters are computed.
■ Peak Concentration and Length
Standard Deviation is computed from GLength using:
GLength y
stdDev y = ----------------------- (12)
2

■ Dose and Standard Deviation

Given Dose and Standard Deviation, the Peak Concentration value is calculated using:

Dose ⋅ factor ⋅ 2
C peak = ----------------------------------------------------------------------------------------------- (13)
 y peak 
π ⋅ stdDev y ⋅  1 + erf ------------------------------- 
 2 ⋅ stdDev y 

4 –2
where factor = 10 because Dose is in cm .
■ Dose and Length
Given Dose and GLength, the Standard Deviation is computed from Eq. 12, and the Peak
Concentration is computed from Eq. 13.
■ Peak Concentration and values at the junction
Standard Deviation is computed from the values at the junction using:
y depth – y peak
stdDev y = ------------------------------------------------------------- (14)
– 2 ⋅ ln ( C atDepth ⁄ C peak )

NOTE C peak must be greater than C atDepth .

■ Dose and values at the junction
First, Standard Deviation is computed from:

 y peak 
C atDepth ⋅ π ⋅ stdDev y ⋅  1 + erf ------------------------------- 
 2 ⋅ stdDev y  1 y depth – y peak 2
---------------------------------------------------------------------------------------------------------------------- = exp  – --- ⋅ ------------------------------  (15)
2 ⋅ Dose ⋅ factor  2 stdDev y 

Sentaurus™ Mesh User Guide 123

O-2018.06
A: Formulas for Analytic Profiles
Available Models Along the Primary Direction

Second, using stdDevy , Peak Concentration is computed as in C .

–2
NOTE Eq. 15 is an implicit equation and Dose is in cm .

Error Functions
For error functions, the basic set of parameters includes C max , y sym , and ELength y and can be
computed in different ways:
■ Maximum Concentration and Length
The basic set is complete and no parameters are computed (see Eq. 5).
■ Dose and Length
Maximum Concentration is computed from Dose using:
y sym y sym y sym 2 –1
2 ⋅ Dose
C max = ---------------------- ⋅ factor ⋅  ---------------------- ⋅ 1 + erf  ---------------------- + ------- ⋅ exp –  ---------------------- 
1
 ELength y  ELength y
(16)
ELength y ELength y π

4 –2
where factor = 10 because Dose is in cm .
■ Maximum Concentration and values at the junction
ELength can be computed from:
y sym – y depth 2 ⋅ C atDepth
erf  ----------------------------- = -------------------------- – 1 (17)
 ELength y  C max

NOTE Eq. 17 is an implicit equation.

■ Dose and values at the junction
Maximum Concentration and ELength are computed using the following implicit
equations, which follow from Eq. 16 and:
y sym – y depth
Dose ⋅ factor ⋅  1 + erf -----------------------------  = C atDepth ⋅ ELength y ⋅ (18)
 ELength y 
y y y sym 2
 ----------------------
sym
⋅ 1 + erf  ---------------------- + ------- ⋅ exp  –  ----------------------  
sym 1
 ELength y  ELength y π   ELength y   

y sym y sym y sym 2 –1

2 ⋅ Dose
C max = ---------------------- ⋅ factor ⋅  ---------------------- ⋅ 1 + erf  ---------------------- + ------- ⋅ exp –  ---------------------- 
1
 ELength y  ELength y  ELength y 
(19)
ELength y π

124 Sentaurus™ Mesh User Guide

O-2018.06
 A: Formulas for Analytic Profiles
Lateral or Decay Functions

Constant Functions
Constant functions are useful to define substrate doping mathematically:
g ( y ) = Constant (20)

1D External Profiles
Real 1D process simulation results can be read along the primary direction. To complete the
2D profile and 3D profile, you add an analytic lateral function.

The values that do not appear in the file are interpolated using an interpolation function. Every
species has a corresponding interpolation function predefined in the datexcodes.txt file.
These functions can be linear, arsinh, or logarithmic.

If h is an interpolation function, the value at point y is computed from an external 1D profile

as follows:

 data i y = yi

g ( y ) =  –1 y – yi y – yi + 1 (21)
 h  --------------------- ⋅ h ( data i + 1 ) + --------------------- ⋅ h ( data i ) yi < y < yi + 1
 yi + 1 – yi yi + 1 – yi

Lateral or Decay Functions

Lateral or decay functions are evaluated on the valid lateral domain (see Figure 39 on page 118
and Figure 40 on page 119). They are defined as the decay along the lateral direction and
depend on the distance from the valid primary domain of the point to evaluate. For 2D
simulations, this distance is calculated using the baseline as reference. For 3D simulations, the
distance is computed using the surface as reference. The following models are available:
■ Lateral Gaussian function
■ Lateral error function
■ No lateral function

NOTE Lateral or decay functions are not valid for 1D simulations.

Sentaurus™ Mesh User Guide 125

O-2018.06
A: Formulas for Analytic Profiles
Lateral or Decay Functions

Lateral Gaussian Function

The equation applied is:

1 x closestP – x 2
f ( x ) = exp  – --- ⋅ ---------------------------  (22)
 2 stdDev x 

According to Eq. 22, the required value from the user is the standard deviation, stdDev x , along
the lateral direction. You can define it in one of the following ways:
■ Provide the value explicitly
■ Provide a factor with respect to the standard deviation along the primary direction:
stdDev x = Factor x ⋅ stdDev y (23)

■ Give the length of the Gaussian function:

GLength x = stdDev x ⋅ 2 (24)

Using this function, the decay begins outside the primary domain, that is, the overlap between
the primary, lateral, and decay domains is zero. Figure 43 shows this effect.

Baseline (window mask)

Figure 43 Using Gaussian function as lateral function in two dimensions

Lateral Error Function

By default, when specifying an error function as a lateral function in an analytic profile, the
following equation is applied:
x closestP – x
f ( x ) = --- ⋅  1 + erf --------------------------- 
1
(25)
2  ELength x 

126 Sentaurus™ Mesh User Guide

O-2018.06
 A: Formulas for Analytic Profiles
Lateral or Decay Functions

According to Eq. 25, the required value from the user is the length for the error function,
ELength x , along the lateral direction. You can define it in one of the following ways:
■ Provide the value explicitly
■ Provide a factor with respect to the length along the primary direction:
ELength x = Factor x ⋅ ELength y (26)

For this model, the overlap of the primary, lateral, and decay domains is not zero. The lateral
decay starts inside the primary domain as shown in Figure 44.

Baseline (window mask)

Figure 44 Using error function as lateral function in two dimensions

In some situations, the previous formulation can lead to a change in the total dose defined by
the analytic profile. In those cases, you can use the lateralDiffusion parameter in the
Interpolate section of the Sentaurus Mesh command file to change the formula used to
calculate the error function (see Interpolate Section on page 29):
x – x max x – x min
f ( x ) = --- ⋅  erfc ---------------------- – erfc ---------------------- 
1
(27)
2  ELength x ELength x 

For higher dimensions:

f ( x, y, z ) = f ( x ) ⋅ f ( y ) ⋅ f ( z ) (28)

These formulas are used only if the reference element is rectangular. In this case, x min and x max
are the minimum and the maximum coordinates of the reference element, respectively.

Sentaurus™ Mesh User Guide 127

O-2018.06
A: Formulas for Analytic Profiles
Lateral or Decay Functions

No Lateral Function
This property is valid when Factor is equal to zero. In this case, the value of the lateral function
is given by:

1 x ∈ PrimaryDomain
f( x) =  (29)
0 x ∉ PrimaryDomain

The lateral domain is null.

128 Sentaurus™ Mesh User Guide

O-2018.06
APPENDIX B Doping Function for Discrete
Dopants

This appendix describes the doping function that is used to transform

discrete dopants into a continuous doping profile.

Sentaurus Mesh can create continuous doping profiles from discrete dopant distributions
obtained from Sentaurus Process Kinetic Monte Carlo (Sentaurus Process KMC). This is
accomplished by associating a doping function with each discrete dopant. The union of all such
doping functions defines the final doping profile for the structure.

Doping Function
It has been suggested [1] that the charge density associated with a discrete dopant be can
separated into short-range and long-range portions, and that the long-range portion is
appropriate for inclusion in drift-diffusion device simulators. Sentaurus Mesh uses the long-
range portion of the number density associated with a discrete dopant suggested in [1]:
3
k c sin ( k c r ) – ( k c r ) cos ( k c r )
n ( r ) = N f ⋅ --------2- ------------------------------------------------------------
3
(30)
2π ( kc r )

In this expression, r is the distance from the discrete dopant, k c is the inverse of the screening
length, and N f is a normalization factor such that the integral of n ( r ) over the entire
simulation space becomes unity. Note that the above function is oscillatory and becomes
negative for certain values of k c r (see Figure 45 on page 130). In Sentaurus Mesh, however,
the above function is cut off at the first zero of n ( r ) . That is:

3
k c sin ( k c r ) – ( k c r ) cos ( k c r )
n ( r ) = N f ⋅ --------2- ------------------------------------------------------------ , k c r < 4.4934
2π ( kc r )
3 (31)
= 0 , k c r ≥ 4.4934

In this case, the normalization factor is taken to be N f = 0.59688 .

NOTE This normalization factor assumes that the function given in Eq. 31 does
not extend outside the simulation space. In general, this will not be true
for discrete dopants located near boundaries.

Sentaurus™ Mesh User Guide 129

O-2018.06
B: Doping Function for Discrete Dopants
Cut-off Parameter

0.4

0.3
n(r)/(kc3 /2π2)

0.2

0.1

-0.1
0 2 4 6 8 10
kc r
Figure 45 Long-range number density associated with a discrete dopant

Cut-off Parameter
The inverse of the cut-off parameter, 1 ⁄ k c , is the screening length. Different charge screening
models suggest different expressions for k c as a function of impurity concentration. One model
suggests that k c is the inverse of the Debye length:
2
Ne
kc = ---------------- (32)
ε Si k B T

where N is the impurity concentration, e is the electronic charge, ε Si is the permittivity of

silicon, k B is the Boltzmann constant, and T is temperature. However, the paper [1] prefers a
charge screening model that gives k c simply as:
1⁄3
k c ≈ 2N (33)

In practice, k c can be used as a fitting parameter, for example, by comparing the threshold
voltage for a large MOSFET when the present doping model is used with the threshold voltage
obtained with a standard continuum doping model.

In Sentaurus Mesh, k c is a user-adjustable parameter and is called the ScreeningFactor. It

is part of the Particle definition and it can be specified separately for each Species.

130 Sentaurus™ Mesh User Guide

O-2018.06
 B: Doping Function for Discrete Dopants
References

References
[1] N. Sano et al., “On discrete random dopant modeling in drift-diffusion simulations:
physical meaning of ‘atomistic’ dopants,” Microelectronics Reliability, vol. 42, no. 2,
pp. 189–199, 2002.

Sentaurus™ Mesh User Guide 131

O-2018.06
B: Doping Function for Discrete Dopants
References

132 Sentaurus™ Mesh User Guide

O-2018.06